diff --git "a/data/train-00008.jsonl" "b/data/train-00008.jsonl" new file mode 100644--- /dev/null +++ "b/data/train-00008.jsonl" @@ -0,0 +1,5000 @@ +{"id": "train_040000", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "169063769ed5d7eeafd172d6ccdc395d2b005f99"} +{"id": "train_040001", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "1225e31973272eb312af4f33bfdbea88d8eb01b1"} +{"id": "train_040002", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c095edccfd191e392c70ec4fe01f75825f2e594c"} +{"id": "train_040003", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "99b31a686dc8111bce5097731b390b3d16d952fe"} +{"id": "train_040004", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "c337a1555ed8b0796ccb49997b2b9a0e18e7c96b"} +{"id": "train_040005", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b3e37a7fd8f2840a691664c3ee3283e5be52926e"} +{"id": "train_040006", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f0babb971f52704f9392cb4455becf808eddda81"} +{"id": "train_040007", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "01b16b75482c36e37468d97c217e91c6b9ea0634"} +{"id": "train_040008", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "17e108ba739dbda2526688b6afca840f3c69d6d2"} +{"id": "train_040009", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8bdbd802fa4d24c7c9fe84291d5c1fa1eeaa3d65"} +{"id": "train_040010", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "29429b1a1662532460d89455ca9a34b930be2378"} +{"id": "train_040011", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9083bde29cb2be7868c4be850d864904b3f42df6"} +{"id": "train_040012", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "28e5f0deaac5bd957e55b94296a4587a5570cf12"} +{"id": "train_040013", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a07f1a2f16fd622aed357a8b8e20a81b66f98b4b"} +{"id": "train_040014", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a7014fc4f4fe8c8b081b9a750f53298105623fa5"} +{"id": "train_040015", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "9b7204f88abcc953f1febc4ae8938917693226e7"} +{"id": "train_040016", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "16ef567abacd23ced6969b7cd15610f0f422ed3b"} +{"id": "train_040017", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "26dbafae46cf979bf1d27fed5033c9a7b0dc8e7b"} +{"id": "train_040018", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "219d790455e9dd4be179bb87e4c9f2bbb0943e50"} +{"id": "train_040019", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "76731075d4eb4e5b69d996eb6f85f8787280b96e"} +{"id": "train_040020", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3e568f3416d49690ded26e78f263a1e941821b6b"} +{"id": "train_040021", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "91b7bcb9e68927178914d101858f88ebd4ca8c46"} +{"id": "train_040022", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5f0240e205520679ef270f4d4af34e9c74742930"} +{"id": "train_040023", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "44af8bb6e5156ea36edd487393e48746a937e7aa"} +{"id": "train_040024", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2752ffc5cc37938a6d9be3aac63b6e135121c756"} +{"id": "train_040025", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7d5287bbed42a8370911237a00f2a3b1c70ef68a"} +{"id": "train_040026", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f015233311b0743aad7e6525ba28293112fd469f"} +{"id": "train_040027", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "aee5fd62cbd2f7d8c31f457b81a7d848a1a27a3f"} +{"id": "train_040028", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c8e398a93c0239f15e87ed8a1efdd723209b2897"} +{"id": "train_040029", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a2f3a485e8ddc57284a95288dee0145fffa9d456"} +{"id": "train_040030", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e9b39cd0c7dedaecd8960273646747197d086c5d"} +{"id": "train_040031", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a4043432b4a1013bf0c944556da35134f872fb27"} +{"id": "train_040032", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ef9450f1fcb77010e96aba3daa09c598f105b709"} +{"id": "train_040033", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "27b61053ecc757d2655a25b85a1a1d96d9c63db3"} +{"id": "train_040034", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e9e477faf1712149c7e8b8ee01018130db0e74cf"} +{"id": "train_040035", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6c804547d12284c0df859cdc3508fa022a154de8"} +{"id": "train_040036", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "74d3a0caec59351db5d136f2763025f8e9a6d563"} +{"id": "train_040037", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2b23455a5129f5a8c83c427a881daa3e008a67b7"} +{"id": "train_040038", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b6212da244ea2304a7953576327f1dd579e57b98"} +{"id": "train_040039", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "84874f278db1f6e98b6b36b314061161cd437f97"} +{"id": "train_040040", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "33c5b6dd3286dc7838e74bc2c04ac7bd6fbccf31"} +{"id": "train_040041", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ef1d0dcf982c7f4cb3bbd3fa2e0ad135214538e9"} +{"id": "train_040042", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b1935fae419de9dbc4e42e924fe67641776af5e4"} +{"id": "train_040043", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f52a71e1c3de1f3e2344929b674181cd2a0e9952"} +{"id": "train_040044", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6706d52c4eced4c3b59525d425ccf958c36efb78"} +{"id": "train_040045", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "25fa0903e3bcf3c1181325fe325fe1e407c138f3"} +{"id": "train_040046", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b0b1336e99ed116bd5f6ae848f7788632ab4e718"} +{"id": "train_040047", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "99f3b7ddecd75016c8bbe52c59e031b183864b25"} +{"id": "train_040048", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0a0d1b32053a63163d38c55ef45483be35c4173c"} +{"id": "train_040049", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6457ae641b66ed2bca0fc03321b007a51e59797f"} +{"id": "train_040050", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5f2ad6c8e5ffd0bc0c06a89cd76ba03ecefd5899"} +{"id": "train_040051", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a9dc23a13722d16a53b816c3f734045dc9368b57"} +{"id": "train_040052", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "abc5e3b413f45018d8e29dc476c2112727b2382a"} +{"id": "train_040053", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9d61ced89efdfd58111f4ce01aebb0b944bfc12f"} +{"id": "train_040054", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a099657211ff03839238f78ba0d8928f40829010"} +{"id": "train_040055", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "61a185156504d158778ad677892b0dd4fbb96efb"} +{"id": "train_040056", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "36d1afb3beb8d5a5f273fdb10be6d38898b1143d"} +{"id": "train_040057", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0aff886fcdfbc43521a801dbc891b815a9b357ea"} +{"id": "train_040058", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "5fbd8fea0efb2ecbf448efa9fe18460951bfb91a"} +{"id": "train_040059", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2bedfbf83787c6c05e8bee77c8e5d6fad768f1b7"} +{"id": "train_040060", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b78206c6f53cd8ba64c186c8178806d115dcbce6"} +{"id": "train_040061", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "93275fd5e9dc7e1e383465f8d076e13329d323a0"} +{"id": "train_040062", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6d3872d7afaf8fbcf82b98d8ba7a77ee63024e4c"} +{"id": "train_040063", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "fe408ab6b3d1a97730dd41cb26aaf4848f47c505"} +{"id": "train_040064", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "94a80f854359d3e8e81848eeee486133f5a794f4"} +{"id": "train_040065", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "944ce600ddf06a66c1ab7947e9642363d359bcfa"} +{"id": "train_040066", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "bb3b166e234a26592d2800407e84e2fd4675d3ac"} +{"id": "train_040067", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e46d9b5398526b6ac63185f672ff8e348391636c"} +{"id": "train_040068", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9b4dafac1bf2bc013420a9ecdcf67e67578d0c4d"} +{"id": "train_040069", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ea5566ab3fb701b105695f6c6cdbca4e9cd0a001"} +{"id": "train_040070", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8d2591c00b58b7079455f89c497050e85cf29d4d"} +{"id": "train_040071", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "399067b0189f378ef805b0364255a53c18e40b0d"} +{"id": "train_040072", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "74b0b01cbf21e872afcb3ed0b71a3926661f0b72"} +{"id": "train_040073", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "12e78e160631b3ea2eb78a4ede0fa09d28d50d02"} +{"id": "train_040074", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "540b290d5d9fc8ce317d17905f2e4db11d90eb38"} +{"id": "train_040075", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e991407f386595ec4531957c9a93c1c084b3aaae"} +{"id": "train_040076", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "52de1f591efc891372511a569437a0a1caf852e6"} +{"id": "train_040077", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1c597cdb234ac747232cf03391bbf9e18f7160e7"} +{"id": "train_040078", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "6615c83843b2259fb2c9be3b97b91deeb3991cc8"} +{"id": "train_040079", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e1ea0a57e58c19ac65cb363150e5b4fcedc06906"} +{"id": "train_040080", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "82d5ab2b744431ccc3f75c7b0e49f888e4f16b01"} +{"id": "train_040081", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "abb2b5742755f0dd5261ac8ff1bd9ec4b032c0e3"} +{"id": "train_040082", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1f6471f1fe30a678cf7991be9c9abce59be20086"} +{"id": "train_040083", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c9e8081ccd358ec331de8ef68d4bcaf13e74c695"} +{"id": "train_040084", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "56446b72529523aa5ccc83f9b616c2ce0957b97a"} +{"id": "train_040085", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "dcce43c44076fcd3ce9002c6100e2722302f80ed"} +{"id": "train_040086", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "88fd610d4bbde333583c79a3aaedb6519717bce6"} +{"id": "train_040087", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dda79a896280adec1a8b31103ea344c5a6873642"} +{"id": "train_040088", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "88d226319056df084c792e28ca1cc47c00817066"} +{"id": "train_040089", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ea5d9b7a40e7a1a48883b97bb30b585705acf3f2"} +{"id": "train_040090", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3e3a54628dcc9b87647a73bba2062a8f75105dd3"} +{"id": "train_040091", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "173d54ed4e75f02d6383930bdaa2104f9b4316e3"} +{"id": "train_040092", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "83e5ace8856e7c4c2af48993ca3e29578aae201a"} +{"id": "train_040093", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "4ef1378c01f5b5cc43728e58794a72b31b3811e3"} +{"id": "train_040094", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3331b3e749ff02d5f4854810440b954bfd7fcc48"} +{"id": "train_040095", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5c918817dea20f00ca7a1242761ecb669ed97b46"} +{"id": "train_040096", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "42d6659bdfeca2a92935fa027993f80d6576686f"} +{"id": "train_040097", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c7e959ea1a6a5a0a4f0c7c243f98b765adc9886e"} +{"id": "train_040098", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "28e5f0deaac5bd957e55b94296a4587a5570cf12"} +{"id": "train_040099", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "08f902af149364d9206cad1eb4ba10d1bfa27e08"} +{"id": "train_040100", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "75f205a378bfdf3c2d2177e4fd56fd74476c0e46"} +{"id": "train_040101", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1a83a3315d1ef7e7c24500b5fab8634306a72ec0"} +{"id": "train_040102", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f36c83a896e9f15d1911818b0bdb3518d4847610"} +{"id": "train_040103", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a9143255dddc132db3c07f16509a47b90a351f94"} +{"id": "train_040104", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ebbcc6a5102de01d8256f16b19352a4f2736d9f8"} +{"id": "train_040105", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "961d0220e8ad017070ff1e72d7632dab3925ce68"} +{"id": "train_040106", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "bb8067a72bbd7624f7333b40a588b69378abb41e"} +{"id": "train_040107", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0f3b997590796c002f8fe61451245525e18e1237"} +{"id": "train_040108", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0f27ea8fdf097a0fb4d1da70e8cbf4b53e7a26ea"} +{"id": "train_040109", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "99f3e5c4876fe240e2502cd662b2a360f77d9f3c"} +{"id": "train_040110", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "262c27b6ae0a18bd3e5400bbac494a6b9fb16567"} +{"id": "train_040111", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8c81ee9f400a939523c22cb63d88e72262beb912"} +{"id": "train_040112", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8add8a60a75552a9d0744abbef9b3e262bea8c92"} +{"id": "train_040113", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4c0a7cc444d049c7dfa4454341dfc0e8b42d6ed0"} +{"id": "train_040114", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "33f7cbb1d6fe8f56a766943de921298c92cccc20"} +{"id": "train_040115", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "60263ebaba7e398ee5751a757fa6c3d311f79d37"} +{"id": "train_040116", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "59a5f0a035a75990af00535dcf976da513eca136"} +{"id": "train_040117", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ff7e221ad09aa4d39d045eda453598af05b55037"} +{"id": "train_040118", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "fe7b6de905ffa2d36e8c28eaa95d8fe05a158b67"} +{"id": "train_040119", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "703c5d49319209b7c91af87a9d8c0ffee27c1d53"} +{"id": "train_040120", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1c7fe43dd2407a84b191eae91d0d826985482f22"} +{"id": "train_040121", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8664751b738c71a395e8b2927a9cfa4b3890ba3f"} +{"id": "train_040122", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5b6e928dd39ad43643a19aa7ce79acb995dd78e5"} +{"id": "train_040123", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e057de2a73be65d644c45829a8d7c6117f5c3470"} +{"id": "train_040124", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d422cdebc17f75442c8137d17b9c526ec4ef6370"} +{"id": "train_040125", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ded7e22e5a678409f69307a5df3eba25e9279007"} +{"id": "train_040126", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e4a2143c0bffee7aaa147bc4245b9a2027b0155b"} +{"id": "train_040127", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "adfb667c13d0f63571cb83fb0d562a68d53fe561"} +{"id": "train_040128", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "36a16d43d511a8e7888b6012c231e7b2c15359c6"} +{"id": "train_040129", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c167a08bc168d0d70a940cced09d6edab10f7af6"} +{"id": "train_040130", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8cc880b5e36a2c685def39cbf59fedc2590917d2"} +{"id": "train_040131", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "48c99de8497ee6bfa1ad169ee77ae01bfacf7d00"} +{"id": "train_040132", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9893c1c3acc58ce9990f3bd4ab69111bedefaade"} +{"id": "train_040133", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "273d656593ffd1fd8b3c9adb5ab65ab31f2965ae"} +{"id": "train_040134", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4ff58fa5065954b6390998592f46df968367eec6"} +{"id": "train_040135", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "abb48dc373e36d89c8c5128a69574a866992495a"} +{"id": "train_040136", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f8cc82ed8243e1ef31fcf0fd36c8c13ecb062c9f"} +{"id": "train_040137", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "489c9d5b976fbdf069f0a8af6ce2b466bcb71fad"} +{"id": "train_040138", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "cec11222e05d33baf1e563d218c601c3118572ba"} +{"id": "train_040139", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4ee1e843e2b182046a58b6c842874b2e0fdec8cd"} +{"id": "train_040140", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9888a6c0606d243a95814d3fc44239c6b3b20dda"} +{"id": "train_040141", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9e06b8376319825fd8269e2839e644c3160fd9c7"} +{"id": "train_040142", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "eb109b535e5356db691f54d5564f7ebe5bf37d58"} +{"id": "train_040143", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "af4a641702e8aee3b28a5d8e2197a01e83a22d6e"} +{"id": "train_040144", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0317b8417a2e623cbeeea9512d5d44aa5bf52574"} +{"id": "train_040145", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4ee67a44f71836f5519c211a8dca96c6f7df677c"} +{"id": "train_040146", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "63862e15d340b3417705d7f6af8d5ee762fbf368"} +{"id": "train_040147", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "82a70b2fba8496778bbdb7a33d8ed37ae88ab4b3"} +{"id": "train_040148", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2346b99f42bf2cd46400d4f273c7ae7bdce5d5b7"} +{"id": "train_040149", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "90fa35deae7d8e9f8ade9eb13c49b3f881be8489"} +{"id": "train_040150", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "50f7db09183e823c2c0d7bd9f8b7e43899770d0e"} +{"id": "train_040151", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "48da2302a2b24d2e8f827af8612456dbf9ba5564"} +{"id": "train_040152", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "fe3dd7119fc96afab48e357475d48cd3c7931d07"} +{"id": "train_040153", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e18fd6f3228ddd8cdebe494bdb0609629727bc71"} +{"id": "train_040154", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e09169b21a81bd641f47799edbd8c75d7adde409"} +{"id": "train_040155", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "12c8dd0fd58e8b3421f5cd4eb87bc4865f9273a9"} +{"id": "train_040156", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8339ddd7043f1f609ea33c506edc5d7311e53c83"} +{"id": "train_040157", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0c954badfddd2d0698b99cbb92920463580f607a"} +{"id": "train_040158", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2ee80fdd3b568575a6013b78948833ad961cdd03"} +{"id": "train_040159", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b0f14b6e73b3f7fdd33d93e9c5442f0eb48b3692"} +{"id": "train_040160", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3e14721dcfd915628a81f972b399eb8da79524ab"} +{"id": "train_040161", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "0b0b2dc4b93f95483c1d22de893efcd3ebc483e6"} +{"id": "train_040162", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "bcd3a470f2753ef468070062d87c4330043d8df9"} +{"id": "train_040163", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "80f8d206109e9b89cc458e69ce7f5f26448a7b0c"} +{"id": "train_040164", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4944c4ce4acf2c9cd176d36d3cea519e56d72994"} +{"id": "train_040165", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8d3abb684cf92f7354c4e3eace4abb735e5b1030"} +{"id": "train_040166", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1538eda49604d8df967dd8d6d07615ee431bf8c8"} +{"id": "train_040167", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d2e2bb519202e1680081ff1b40597fd61568c3a7"} +{"id": "train_040168", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "970b22c3c1eb2888b0412236e0fa4e660f979d6e"} +{"id": "train_040169", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "96755d0cc1875c963c583cbd88ae6102994c9e8a"} +{"id": "train_040170", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cdd2b25aeefdd6f5099bb1645e2c76691b7fcb63"} +{"id": "train_040171", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "748cc9644ef5615396c4b9800198ef75f7cf2c06"} +{"id": "train_040172", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "72fbc3725da4ae9e1260a0be47328e80477645e6"} +{"id": "train_040173", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e188d252620e50bb11491a8bab76815164ae2a54"} +{"id": "train_040174", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a89f06bfc8e73a42880f0611353d8d728f7b31d9"} +{"id": "train_040175", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "648218a315ad11a991c55b8c7bf59b6febb41132"} +{"id": "train_040176", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "10c83d3941bf3f152da3ef0da8cc35f0e9c17501"} +{"id": "train_040177", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f3100847edfe874ecd20d5df8c5c0b8759420d21"} +{"id": "train_040178", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e9a6fe4ec974085afa491514686f2fcd29c95ad8"} +{"id": "train_040179", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "62f1b6f4a460bf4a742d7a839f9a69c1cd3937fc"} +{"id": "train_040180", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "7a8b5beeb1063f6f2e6e06a9a178ef435676d10d"} +{"id": "train_040181", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "670482ad4257ee4b3867e2900363359d08cdd025"} +{"id": "train_040182", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "40307d066c9119cf37dff820e04cf6d25de25e95"} +{"id": "train_040183", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c0ab191a0027f385a40852d791b55ce370f5f4b6"} +{"id": "train_040184", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d7e02ccda21533a39e6b05d932de08220a96c123"} +{"id": "train_040185", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "088fd0f158b6df203ab56ccd3f93f31a3a3ec51a"} +{"id": "train_040186", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6e5c7aee82186c31c9d74e0abdc2870c0c623921"} +{"id": "train_040187", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7785ea2e8277b140ddc68bb39de684c2bfed3a99"} +{"id": "train_040188", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4e33d7e961ba836a29161065ac56315d61dc173b"} +{"id": "train_040189", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "98fa5353bf2b2de9c3006907a0a1799e2a0b45c7"} +{"id": "train_040190", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e079fb5ec96bb8d0ee46857150344cbe693b96a6"} +{"id": "train_040191", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "07083013a4241404ef6327746d239a387627a967"} +{"id": "train_040192", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0073d9f517528c565ec725ac49f21a6e11dd687d"} +{"id": "train_040193", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "dd8b3937e4afcf7d6ff0c8c3ecda974d16ab697e"} +{"id": "train_040194", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "83cdc50939636702f4301be3f79cecea73846fb8"} +{"id": "train_040195", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e164159a72beda78681c9ea9e9efcd54a572136c"} +{"id": "train_040196", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "4ee8eb10bafcc22cfff35be36f70c50354df7024"} +{"id": "train_040197", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "224f43f3dd065b285f1980b51e2fe9949e51eaf5"} +{"id": "train_040198", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f18410317464e1b58b7d50627cb98f8e52186190"} +{"id": "train_040199", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "586e6f92f9ef2e1c7f5ee7ac9bfcfcdd80dca82a"} +{"id": "train_040200", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "697eaf47cea15e6783a980363e82208c91db4969"} +{"id": "train_040201", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a6807096b568ba16005187ea1fa520009a1cd893"} +{"id": "train_040202", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e6099f69bd4920a54f067f67cbabf7484c347b11"} +{"id": "train_040203", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1e9689e3436da2fd8dd6df896790b153e1544c89"} +{"id": "train_040204", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "bb1d8a322e789715b8f8e674ab4473bb4d155367"} +{"id": "train_040205", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fe17c3b3f3cd10f4a7871b9ddba4243cd059da76"} +{"id": "train_040206", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a7874947e85e360e7e4bcabdba7af3fdab6b69fa"} +{"id": "train_040207", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "cf263be34b49e68bc6312d884bde6960eaede738"} +{"id": "train_040208", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "cba47cdd7ea57a93948c5e229adf8fc9b847cafd"} +{"id": "train_040209", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "4d182bebeeb883bc8852c9a4d11a8d1bc974252b"} +{"id": "train_040210", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ecef79f3366d0506614ae2510b730e5abe4eb686"} +{"id": "train_040211", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "38cfce528466546ad2e8d4d84872dd9ffd7e871d"} +{"id": "train_040212", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2147cfaca21582587128846374359883238998ea"} +{"id": "train_040213", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dff3cfb9dc0aafef3f15ba0d8c736205cc6d9475"} +{"id": "train_040214", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ffb8509f7ed2b9df601866db9cf50ab421f32853"} +{"id": "train_040215", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f4df679c7ba202de10ae072a05117b8d2f57cb8c"} +{"id": "train_040216", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3a14f4bf37a10ef2988d485364192a13b47e93a8"} +{"id": "train_040217", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2abb2b077b48e0447c3086196c02a3b6593736b2"} +{"id": "train_040218", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ef0ff2783932194e58863b59d2222eb4b3b13deb"} +{"id": "train_040219", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ffa2ec4f10bf3de1bba763aed946558fe8de462a"} +{"id": "train_040220", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "edd44bf9b144f85cc16e19fe8f5eaa62e022f10d"} +{"id": "train_040221", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "271c641dec6e1854f92510e5abca7a48bdf1dc26"} +{"id": "train_040222", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0c6040899743c15d836f492c12dc5abdbf673121"} +{"id": "train_040223", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "106a15cb61098d201ae5c98bf4e76f06005de80f"} +{"id": "train_040224", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "aa06d2e4fa8a3658cbc9bcde2ac1eb20881d50fa"} +{"id": "train_040225", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e8bd291339956c936bce6e95c646592253a861cf"} +{"id": "train_040226", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2e8031bfddeb26c6a5e94244db964094965fb837"} +{"id": "train_040227", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "96628281e5333dec2e9c5d6d10406ee4ef4c2fc2"} +{"id": "train_040228", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8d47e6c4e03cfbb2132981a6b743b59b8e3b8953"} +{"id": "train_040229", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "438dabaa2dde99ab9225ac826816d34955ea6099"} +{"id": "train_040230", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d43387e6ef70a660f45a0cc449cff579c035f969"} +{"id": "train_040231", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "7630dc8aa4f8e0884286fe3f45657438e3946c6a"} +{"id": "train_040232", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d4224f7210d9467d5fcfb806a7c307ff929bcd7c"} +{"id": "train_040233", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "26e50be306689d348d6dd4994d5b184d7a69ebbc"} +{"id": "train_040234", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5feed6e6e741c82e49d5e6cbab9a1a6b97d9d6bb"} +{"id": "train_040235", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0b59ec71592723259a9c69728ba5b2790617fc59"} +{"id": "train_040236", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7bab7f1eed65525fd5af9a30d0a2810daf22f709"} +{"id": "train_040237", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8a0b9e2c6c4a1d9cfb51b2f8c008be1242625128"} +{"id": "train_040238", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "98d813860002845d670e83d810902f45953009d0"} +{"id": "train_040239", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "65dda80bcdc4f31fc57153747e0dd7bf6d12a3cf"} +{"id": "train_040240", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7e09c1266f611024ebf089221e3b97e36675eeed"} +{"id": "train_040241", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "fdb6ec7c9a84c46f01831e20d20bef81dc769d9b"} +{"id": "train_040242", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a673fc3910f348c4481ce8f164949dd95679ffd7"} +{"id": "train_040243", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "027400c74f50707388a83706bfb7c3840d123f8d"} +{"id": "train_040244", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "bfa580a39ae2cd09dc9fbf2e44faaa548633f775"} +{"id": "train_040245", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1b858b4a13c1b915b47be7b1e46be4b82f627c45"} +{"id": "train_040246", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "094a8412e26f400f1fa42c2f4960319ed4abaedb"} +{"id": "train_040247", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b898676fe392db7374e3eb75d2087b1b60191c1f"} +{"id": "train_040248", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4a4739615342983ec2878396b25cbaf14a4ad858"} +{"id": "train_040249", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4898cf3e38bcec89b1bb91297c92dee10dc46651"} +{"id": "train_040250", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8055cb8908f27e4fee355e7d1d1223d3a39274c4"} +{"id": "train_040251", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "01b888c77840d2684943835b83fee7a23b64e88b"} +{"id": "train_040252", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ee783e2f5d81cb7e49c9b527291b343135c6b3e9"} +{"id": "train_040253", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "91061fd2777a68f78c13bb3ae15591eba2e6d790"} +{"id": "train_040254", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "89f024b77d96f18ee229a67469d09e22375405d9"} +{"id": "train_040255", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0bf2373ab01ce8c36a05f6c1b08f6c9683e95665"} +{"id": "train_040256", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "960a8f6913212a5b178c1b214a87d05c93257723"} +{"id": "train_040257", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "58bc280d3a8d8182f86aaadc40c13e593a7df6a9"} +{"id": "train_040258", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "bdd72856597bdab9f988d7172e209fbfbc6da6ab"} +{"id": "train_040259", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "282cf5177c6ae3925dfa2506b89434eacb9e6564"} +{"id": "train_040260", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "69f423e27f5bf00a7c84bcc99337aff260efd50d"} +{"id": "train_040261", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "027eb0d7c4e9f4e171be0ca359c8b211a2338ba7"} +{"id": "train_040262", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3bafa409342d4f3709c3a07cb0a30d431c08c6e8"} +{"id": "train_040263", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2219f274e784be672a18e9705f9fe138f7969530"} +{"id": "train_040264", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "cd8f95bf8042c11d03f238523b7c1623b7d853ff"} +{"id": "train_040265", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0dbeff7bc9991f66f7aa361c910a68905fc0ed46"} +{"id": "train_040266", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d83e7f88b548409404a6d419cb14a1c411f82dee"} +{"id": "train_040267", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "bc943e683993d98b4b55781c57c7bd34da0e3000"} +{"id": "train_040268", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e4b1f098e79f15c58cb530b74d476860d435e0f5"} +{"id": "train_040269", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ca568902b61ffcb0f1e46aceb37605072487ece2"} +{"id": "train_040270", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "9b2c32b8e22715e5a87ed2b12d03a56518eeb890"} +{"id": "train_040271", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d1a11bedfea6827f284bdd72767036746e0200f5"} +{"id": "train_040272", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "58b997dc74b6042dbdf64a02e755b5503113ef5d"} +{"id": "train_040273", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8c81ee9f400a939523c22cb63d88e72262beb912"} +{"id": "train_040274", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "44dd500ec53412bdf21d16949f2608ef20b25494"} +{"id": "train_040275", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3d466372c67a8210216d71ee576239c925ddc60c"} +{"id": "train_040276", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f5d0fd588dcfd2bcab65e4b8c474e419d21d0242"} +{"id": "train_040277", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "df3cd0bf14f7255139b07fb89beb35a3a8f4f300"} +{"id": "train_040278", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8ca9ed843cfa5fcfa607fa11369ad7c29dd32456"} +{"id": "train_040279", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "400ed9801dd4fdf8bfb90082a6cf760027cb1421"} +{"id": "train_040280", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "9d2eda413bcada2da67c3ad8b76eb9c07f235fbc"} +{"id": "train_040281", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "29e19804b05f616f785d07e17f13af20b468a384"} +{"id": "train_040282", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "092cf2cfe64d520e9831b44b9c0eaa72a870d4ee"} +{"id": "train_040283", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "63d4595d40d5b6b1bccbabd637108adeb77997d4"} +{"id": "train_040284", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "92518a3a15c6f240543a48497cf7d182bd799b7b"} +{"id": "train_040285", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e512cb7d3bbca18c722a2e5435d06f2cd5a583ac"} +{"id": "train_040286", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "cb33da329f879e86340edb5dca155bdc67101409"} +{"id": "train_040287", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1a82a1e5286b174a8b3afb5df7f9e865e542e4c8"} +{"id": "train_040288", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "fb5c27eda16b2cdbc5428931c4b8865d38adef49"} +{"id": "train_040289", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d52fcd70276914190a34bad33d14a86f46598260"} +{"id": "train_040290", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d3eec790d19022d14ca6b422b303aae5c2a86f15"} +{"id": "train_040291", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "fca1d4dc1ef6f82bbfc8057eec3825754077227b"} +{"id": "train_040292", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9e033604c6dd9d4fc235e9a18432c706f82c2463"} +{"id": "train_040293", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5bc762f153589aff60f5b8105c72ad7d98256507"} +{"id": "train_040294", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9d5ea1656c82925060ee837f4ce75844fb6c42fe"} +{"id": "train_040295", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ffd188ebcb89435baa962b7937aeaf9d802162d2"} +{"id": "train_040296", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "bc67025d217470eaaca327f327302eb5e852b4bc"} +{"id": "train_040297", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e49ea2f2dfb62099fcf5397299bc57b467229c61"} +{"id": "train_040298", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "261ec6908c75a86f940f21f4625e942be69880f5"} +{"id": "train_040299", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b35dbbc5b517fb701256e0d7fb62409a8f8f4f96"} +{"id": "train_040300", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8faf464a0d07a462465a2013aa06ff2787dc5c37"} +{"id": "train_040301", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ea615adfdc9c6ee3202725ba8f360f05fe23bdf2"} +{"id": "train_040302", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "967becdc58aaad9e893b60d52cc1be5e695fdf16"} +{"id": "train_040303", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a659e7e923c4f470c12f510e51c3949c3d64d07f"} +{"id": "train_040304", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e84aa149adc67aa6a9f9628161aea034544738a3"} +{"id": "train_040305", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7fe2c217a2b88bdf10c30a418b2540e50ecef71f"} +{"id": "train_040306", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "20083b3d4ba9bad3d23ada4be47545e19dc24e13"} +{"id": "train_040307", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "5e129fdfae7424e6d5ec41297cb6d3f655883834"} +{"id": "train_040308", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ce94ef50fa10b2f3aa7c4e552b9006ee80d45465"} +{"id": "train_040309", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ac4781e10faa3b1ce6bb4ea2fbdc10665191061d"} +{"id": "train_040310", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a7a1a21251736991bc0dfb27ab9a11f3abdbc384"} +{"id": "train_040311", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3468c3744911168c82afd6c20ba9b513976b599d"} +{"id": "train_040312", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "67b32b23e86cf645635f4c47c2d789fcb6e2d2ae"} +{"id": "train_040313", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a4e8aef6754fa07697ee9735c3e3948d9cefd40a"} +{"id": "train_040314", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f5ce2463538063b6667275d604cb359bed29d04e"} +{"id": "train_040315", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9d6cf952a2c5f52adf99e07350933c8c8813e4b3"} +{"id": "train_040316", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "03319c304a41f59df3e6dc99a889d1dc1f394566"} +{"id": "train_040317", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "808585028535c46c723f3773812e6383d93af596"} +{"id": "train_040318", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7e46d3934818a3a1e8cf939986f1c5bf7704a0e4"} +{"id": "train_040319", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7d72fa70eb3f949d0eec909695c96a9676b940c1"} +{"id": "train_040320", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "703dd8161c879658aeadff332ebebfc45afd3449"} +{"id": "train_040321", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "11c78b041c6bf4c6f893fcdd23b6bcb5b8093482"} +{"id": "train_040322", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8cc9fabb7dc77b6497bcaec029cb5ae00f842587"} +{"id": "train_040323", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6433c6d3ff94516284a4b5e5fbf74a799589dfb4"} +{"id": "train_040324", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c15734b3fece618c3f8455c20eed595e50317856"} +{"id": "train_040325", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "43c3e2dbec0d26c38b59f3e1d231bbdcc9efeeb5"} +{"id": "train_040326", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "244d7cb5b2fda6de457112bcb252ebb7e3991155"} +{"id": "train_040327", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3ae4e9a0d3675ddda147282cb4120508f0428afc"} +{"id": "train_040328", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "1bec3c9145d459f01827511a63145fa91ccd645a"} +{"id": "train_040329", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "47dc9abfb3f02aee0787d4d00047f812f70f55fa"} +{"id": "train_040330", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f796ec0490ee3ca83cb943d3cc747f1b99f20be1"} +{"id": "train_040331", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0511033023441f519ba6545d78de82e5fd688b20"} +{"id": "train_040332", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8e5e2c00a8ac5f70002355608bd14808e8f6edad"} +{"id": "train_040333", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a8ea131e080f89c48c0b13a7914f9d44a82c1855"} +{"id": "train_040334", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "775737078edbdab8936bb7f56f1a4caaaeb893b9"} +{"id": "train_040335", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e252a8b8f9ec096b8da5a33c584c4ebf4351705d"} +{"id": "train_040336", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "45e13dcdcb1d2183ede6f6102d22ff1910844214"} +{"id": "train_040337", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1e6728392e022127590cb433e26afd3a2d0bdfa2"} +{"id": "train_040338", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "09ab1074f3e9784b40cdca8b6b56808cc5ac4aff"} +{"id": "train_040339", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "56a1ff83897c9e326a7ef5f8d52f0017f50ad18f"} +{"id": "train_040340", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0984e349f86ea65042f6ea51aa9e95c434ddbfa6"} +{"id": "train_040341", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e27dda8422c74bd772a791a9bc55581e52e038c5"} +{"id": "train_040342", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "825a2aecb2ece9de399d80ed8c7f8a3839bd66f7"} +{"id": "train_040343", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "37c7a971c79a067d2c3ea62ef91b24c8b127142f"} +{"id": "train_040344", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2d29c685bcd0f834498117bb39bc635b1c3422ba"} +{"id": "train_040345", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "398ecd3dcb19ac1cc6bd8c045ec3bd1c2935f6c7"} +{"id": "train_040346", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7954aa1198e17a0a7fb67e2c1761569798a0deaa"} +{"id": "train_040347", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "78a6bdfcd1667a64fc7bf488f4ca05f07793171b"} +{"id": "train_040348", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9d0ac12eed1779e9eae63151c1e72c02470642b0"} +{"id": "train_040349", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ef825dc6a26bd8cb6fd55c95db410f499229a13c"} +{"id": "train_040350", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "71b81014f75723464196bdb55f40c3927f7fc2a2"} +{"id": "train_040351", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9ec0556e6dc16793a8b47e7bdea32dfe71810d38"} +{"id": "train_040352", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2eade9e575b80b2d2c853969d9939c0c974bc880"} +{"id": "train_040353", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "291a62a253c0098699356a09cce1b517ca105ed5"} +{"id": "train_040354", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "83a9682d39f3c36bd979a59cbaeac06e36f4af65"} +{"id": "train_040355", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "07162dac8d7db90a706c2b209cb84884a1ba1aa0"} +{"id": "train_040356", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "dfa9d8b937fadecf2ccd38b4ba262b092ad8f7a4"} +{"id": "train_040357", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "cf58bf578d34f85e426b415a92ddece2c4f5f085"} +{"id": "train_040358", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f0460bc2b11664aed56ab1bdb806969b6a4a95b6"} +{"id": "train_040359", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e9c10dae677500fdea99579c27533dec017b1670"} +{"id": "train_040360", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a6030456ad994b98584273302e2cf585dd9eeaff"} +{"id": "train_040361", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fe130f695827b856942d81033e319f0e0c54e0a4"} +{"id": "train_040362", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b7ae29c1ab0235a52c638c811c0f4934aaefd479"} +{"id": "train_040363", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a0ed3ea660128ae22bed208168526b1c90bffe62"} +{"id": "train_040364", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "3298760228bbe4f94eabb0a212b34992166bfafa"} +{"id": "train_040365", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "4c6f6c1edc77f0f4b880f160f2d4192421d42d31"} +{"id": "train_040366", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9e59f20082931f76b936e49794f1182ddb66c1ca"} +{"id": "train_040367", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a995de2e883bb14ad9e5beb48b61353e56c592d3"} +{"id": "train_040368", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8de3f3861ba4e2d0fcf9b0285768e78927ac7b48"} +{"id": "train_040369", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d9242ee975f7bea792ba1ddaf4821921b1d8535b"} +{"id": "train_040370", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2b6d4d820d32022f1f9d26751c4e8e1c6ba32294"} +{"id": "train_040371", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5ff8a55f5217e205ba33e2d57f77306407c8d0b4"} +{"id": "train_040372", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "65a5e77374b984001aa1ed2abbb976e3915bcde4"} +{"id": "train_040373", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5e9b6a1405053463a1d526ce3e3a22a2524ef273"} +{"id": "train_040374", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a27a9984e1027f740197333a849deb629ddac50d"} +{"id": "train_040375", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d8ae3168be15c6d8be50c93c21710c8720bca542"} +{"id": "train_040376", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a4a1024b81cbc08e3452cad998bb8eeefe536be6"} +{"id": "train_040377", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2529313be57e05b89f818d015b191a38a32a517d"} +{"id": "train_040378", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0ce2d3c62e32f67774c7c812c8190562655c575f"} +{"id": "train_040379", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "bd9967c4a003e341d1715a0d220490347b08a065"} +{"id": "train_040380", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "3042e8d8327d198fa46a2c920aeef3df47b5fc09"} +{"id": "train_040381", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0ba07aa5c86f00bfeb9e2e74517fd07354a067e3"} +{"id": "train_040382", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f8289300c623cd2281fb8c758407aaff2ba4f920"} +{"id": "train_040383", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d0f2f79aca34383611046a65189e09a760f16dc4"} +{"id": "train_040384", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "36f6df00cd5f3bbbdce7d36310fedf1c109fa55b"} +{"id": "train_040385", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b24a6cd1c9da599fa37c4f965aebab56e074298e"} +{"id": "train_040386", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d380948325759dfd32a0c35d205a510215cfb7ec"} +{"id": "train_040387", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "2a55fa20ab377a986b29071ce18703127f04baf5"} +{"id": "train_040388", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c66579d87f36b2a02f831e0c728fc45a1f712734"} +{"id": "train_040389", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8563974e1995211ec26b25344d72bfb91e489186"} +{"id": "train_040390", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "30bed7470393ea0b1e5488dc263c2bc92d7382ae"} +{"id": "train_040391", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e80d7f0b7c33fb844c49dde41ab0571c78aff4c9"} +{"id": "train_040392", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "32c15d839362b4c15a93f7fc8658e1fbf6541f7e"} +{"id": "train_040393", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8096329737f7ef4155874a28a5d6e1a8a19fa90a"} +{"id": "train_040394", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "fe40a5ccc84c09bf38e0f85edc5d4ba5ac20cc7d"} +{"id": "train_040395", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b7676cbf0451df26ed6daf718d18e1b181a1a95a"} +{"id": "train_040396", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "05cf77f4da3c8882c1ff212715279a43a00eaa45"} +{"id": "train_040397", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "94f410217b7e5c87aaa0bdf3fea5717b84c54790"} +{"id": "train_040398", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1aeca1fea86d07a2acfb1cd86b403ca0d946351a"} +{"id": "train_040399", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0ced0b59a01e204ba1063482060b29b7eb06d257"} +{"id": "train_040400", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "74b1b5f766a335a9ead6222010aee73a4e45875b"} +{"id": "train_040401", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5e568295c03f17cefe5845316120b69868990146"} +{"id": "train_040402", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "94a80f854359d3e8e81848eeee486133f5a794f4"} +{"id": "train_040403", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e0073c18d96168a2c25b4b34112b9374c556a89c"} +{"id": "train_040404", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dd64063e9de4a02e9873fddf7d00803949da0623"} +{"id": "train_040405", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b325d27cc9ce8d8c718e8d692bdbc08491275fdb"} +{"id": "train_040406", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3860bfd8641d09ff68f6cc68f2f9b21e7178d300"} +{"id": "train_040407", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "09bb1166de5a5e98d1c62100ffdffba31a4ceb18"} +{"id": "train_040408", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e30d6413a5d1e0b839ace5bca735d7772e8a50ed"} +{"id": "train_040409", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "4a9dc9f6082fa526199e79425b8261e8c2a9c5f5"} +{"id": "train_040410", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a7fb6fb23e639d23cdfe5599b3ba659560901da6"} +{"id": "train_040411", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "624a45b07227ba6753988ddcc50d18204c883492"} +{"id": "train_040412", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a7d8e2b4509395c004b07f364aa8e6ace794e93b"} +{"id": "train_040413", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8bab3fc4a5bab48112c4d67554362e922fb23ce4"} +{"id": "train_040414", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "61f954bca15299274332459e7b66f12b67679e7a"} +{"id": "train_040415", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "16295dd47e33c1a083935ac043c3f3434a60e4a1"} +{"id": "train_040416", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7c79a0ee144dddc42076f41648575e0bd84c360f"} +{"id": "train_040417", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ce9f21ab81a3e1f1bf7cd25f9c6e5d87c03ace5f"} +{"id": "train_040418", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "678ae6324e32b2ea92e1005605a87384f785fa4e"} +{"id": "train_040419", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5cde7b691ccb1c5c7f2aa5cb149aff43cb64ce48"} +{"id": "train_040420", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ca46a3ae1a620460e94d42e8015eede3e93e26ee"} +{"id": "train_040421", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8fe10d93c943c5b6711a1aa9c27c4d2597cdc205"} +{"id": "train_040422", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1507a709d1c4233dd981087325e10b0d21e4e8c1"} +{"id": "train_040423", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "416d88e4289a63131e64810458474864a6d275e5"} +{"id": "train_040424", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8316fa8633082ec7dbbf8a299fb6506c6d71d419"} +{"id": "train_040425", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8fc8d6616db2578eccb103b58a88af1c0c035e1b"} +{"id": "train_040426", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f0e343506c09a8b97d6abd300c867a901dd77d91"} +{"id": "train_040427", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ee0474866b1444d06b146437cacb363211e6f03e"} +{"id": "train_040428", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b038dcafb689fe369f36632db9e7dfa883dce6c1"} +{"id": "train_040429", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "404f0bb8aeba66a014c3060fef83dcd550a33b23"} +{"id": "train_040430", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e0a33d8b7c16b7f2a206f8cdd15da982638e85f7"} +{"id": "train_040431", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9395303bb401e1fa151807da48476e2e177209eb"} +{"id": "train_040432", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9cb8ae02407f09b6f1af8b7929ad0f1da5e12809"} +{"id": "train_040433", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a75f4854534bd9eb911335ca927fa1768c8fe22a"} +{"id": "train_040434", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1cfdf247a49c7c50d0ef9592bad0e5f618d22f11"} +{"id": "train_040435", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1500640766415c5c43ff74dbfd245cf4602dfd19"} +{"id": "train_040436", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3e08d3d695019c4ceefb69981481dbeff4d5f78e"} +{"id": "train_040437", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "78ebc25cec94c8e2d5a28dfa3a43a29a2832f7dd"} +{"id": "train_040438", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e98fd2a249a903a93d924659ef7aaf4f1c9d7fbc"} +{"id": "train_040439", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7320c0a6bbfda07ce5ef18cc5151b5e04af9c850"} +{"id": "train_040440", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "22290eb88faef093dacd284d70570960e8b93674"} +{"id": "train_040441", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "9cf62812d6b4945192ea65a7e5d099864b978572"} +{"id": "train_040442", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "616abb073df88ae0d57a9735d64cac98ae7558fd"} +{"id": "train_040443", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a11cdfb44cafe8d50bc471f464e5ff96c13da2f0"} +{"id": "train_040444", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f6558b250535c5f8733c5d4425fde08f8ccfaa57"} +{"id": "train_040445", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e4d61cf3bec1927ec9b1699253445078600c08fa"} +{"id": "train_040446", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "32c63f69bf9a5dee88199dc4889c6b73f8733786"} +{"id": "train_040447", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a7d1933a796cb1cc36f3228181ee9fa291d4d157"} +{"id": "train_040448", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "177bc0df32a53fefe6fba2ba644bed87e73dbd35"} +{"id": "train_040449", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "576bf73e55bd2a3dafaa1c3929c2c61b1bbd8b28"} +{"id": "train_040450", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "001f512ca9d543b81928b93f51f1d4ed04833994"} +{"id": "train_040451", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8b5fc39441792fe74d5b415879869694e9909d82"} +{"id": "train_040452", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2490dbef629598c6732995e30a1e1f48f2b25bca"} +{"id": "train_040453", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a8ceac23746c9cbcd6b8133a880b64561ae8479f"} +{"id": "train_040454", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a8cfc8949780908c16cdf75df10eb62176959abb"} +{"id": "train_040455", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8621a6709b5a5ac5f48d1d12fcb3c46645e0f35b"} +{"id": "train_040456", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a6c6f95af1df7222b17dc50daf8dd8a675ee5d0f"} +{"id": "train_040457", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "20a4cca2daa874407ecebf065d3380f07406fbbc"} +{"id": "train_040458", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3501f9ce735bf72aeb5904beae896a5f6fdd9abe"} +{"id": "train_040459", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9d1bd38b0228dc20d74698908f026b35efe5bcbf"} +{"id": "train_040460", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "73c0cb3fe03d0357d3dcb6ed5458abe89f54cc81"} +{"id": "train_040461", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "437b9a0984fefcec057d05207211b268af16fb5d"} +{"id": "train_040462", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ed125307f0f99b71c5479f8f290fd5804f3f0b60"} +{"id": "train_040463", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "62a2a2ffcac93c9a45f399d2aafb86e2f545af2e"} +{"id": "train_040464", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ba1cfeb911de521e726a097ec22cc1784d46f916"} +{"id": "train_040465", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "00739911e10817bb54722633c95bdc0cf69396ac"} +{"id": "train_040466", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "75f205a378bfdf3c2d2177e4fd56fd74476c0e46"} +{"id": "train_040467", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b35cd65f325742a3744a181c2a22b594e714477c"} +{"id": "train_040468", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "64f4ec55a8ebf8dfd4beb5240f4e0a0eaecbc20c"} +{"id": "train_040469", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0ff36caef1c9ea3e8741c284611af0208d3f8717"} +{"id": "train_040470", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c0451aafc3674dd3a03a7ec49cdace0b5e845dda"} +{"id": "train_040471", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f91b502f3222da8ede6ddd8f6530651e238803d3"} +{"id": "train_040472", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a7321009bb0e64c0352bde7a3bea949e7ca1c24c"} +{"id": "train_040473", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "26d660a6626471eecc052a5127f0f4b6c7e87f22"} +{"id": "train_040474", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ee47ac5a9e6c2de3d221a214b3d94c93af37e65b"} +{"id": "train_040475", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2747f4051adc81b55d7adbfddd4f03864409e192"} +{"id": "train_040476", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "fd5db821d3d08e6df095a6ad92685ca99ac6779f"} +{"id": "train_040477", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b1f957dc554a68cdefbc5ae00ee5be2be3fe7835"} +{"id": "train_040478", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2e8b918418eb3bd5b048f24cb641d53c3e159dfd"} +{"id": "train_040479", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "38874967d852fd2c66ac8d8cce50f95208144bc2"} +{"id": "train_040480", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1cc71091134498597c0104778ee74c5256f72e86"} +{"id": "train_040481", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "226c8191b0f59af76ee3790cf246557b9b0b48ab"} +{"id": "train_040482", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "934c99875848e8f328aa75c0a32fce692821f863"} +{"id": "train_040483", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "24d5d6ce03ed97bbb5def017c37c064cde21f074"} +{"id": "train_040484", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7dcc407b97997a41df05e0c4747720e7fae01104"} +{"id": "train_040485", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c8e8462bbbc5ca142e6ed6a99be285a59ab3b0bb"} +{"id": "train_040486", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7cc21443b619546dc0fb22f81233ae3e0518b808"} +{"id": "train_040487", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6885053916035f23b2dff4cbf9341d260a9b986e"} +{"id": "train_040488", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "90ccd49a2bd900e9ad8757afb07fb99b50d4364c"} +{"id": "train_040489", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "46a6d0f71216c2e7e22301bee3ae81de9706c8b7"} +{"id": "train_040490", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0ecae4fc4bf3b38fe199406f6b20f96c386d8258"} +{"id": "train_040491", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "751d5b9e2965825ac565726e25709baa9df614d6"} +{"id": "train_040492", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "fe2bd2e1ba29d6e619955de07be50574c5f5962f"} +{"id": "train_040493", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "afec30f18a9c74767f1860c531172b19f2786134"} +{"id": "train_040494", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "bd948789a5cba3b1ce2b0d52a8822d374f82d2eb"} +{"id": "train_040495", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "03b4d05e85ed0d41d5f2b8582f416ffa2a65147b"} +{"id": "train_040496", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0d2f3b8f764b6c7110d0387c3a50cff9fc43ec60"} +{"id": "train_040497", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "5bce8830d9344a028b3d6060b47a93743a9367cb"} +{"id": "train_040498", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "6dc3ba37b5b4f2b0df7c8145fd6a8b2b5e718d57"} +{"id": "train_040499", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "356a007ed009caa548d54026305242dd39816fbc"} +{"id": "train_040500", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f03834f21357e481cd8d290c8d91d4b9aaf2a1f1"} +{"id": "train_040501", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8aaf01fbbf3b0c9d41069482f6ba7df16727144c"} +{"id": "train_040502", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f91073b8bb1df2ded60db1de238aed47a1eaad9a"} +{"id": "train_040503", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "710879a07e2e5af5f6b029d4285898e3f6267423"} +{"id": "train_040504", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "01a8d061a65d3a5e63ea994ad16c07cb3a7279c8"} +{"id": "train_040505", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d3a41ed8aebac04396166a4043a774a40001d7c9"} +{"id": "train_040506", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "e1ca50fd7242047e5d8be1918b10219237c7d2dd"} +{"id": "train_040507", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8d1c9bfea729d66e13b61744035d1284c1a63506"} +{"id": "train_040508", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c8e5dca474019039b385e84b1ecf51d873f99b4b"} +{"id": "train_040509", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "877a36c8c2e02b5d9553a775879df51c3f5c79e2"} +{"id": "train_040510", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9018a58bea19bbad216caa5a7ffdf837d05ea6d1"} +{"id": "train_040511", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ce34716cb54cb26a20481625aa36ef42d5436299"} +{"id": "train_040512", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3aac82a2f3cd5e9624bed581e9c69929fb25ea02"} +{"id": "train_040513", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "feadd743ae6ab67f675d2a7de231d9ece2ca5af2"} +{"id": "train_040514", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "871292e406ca2d3bcdbfd24722ef592f1513c515"} +{"id": "train_040515", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "1316cf7fe2ef7ed8ae1d5d93f1a3e9038d9e85b1"} +{"id": "train_040516", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "929496d45462bb0c172f157a40a35434176ee8ae"} +{"id": "train_040517", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ea49fcf5ea93dfd0fe8956b3986cc845fea5f7b5"} +{"id": "train_040518", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "afebfb889043549eb2d20f1eff15b29913184e2e"} +{"id": "train_040519", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "58a6ac16bb85610c0fa4ab1a3173d72c3ff3c8c7"} +{"id": "train_040520", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "02561af755791af61118326abc59ef9c13e56643"} +{"id": "train_040521", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d2e1d3873196dd250a7c2470daa88e8b10f50c5d"} +{"id": "train_040522", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "85a394627d11fb90c2adc6515da15390bbb41ceb"} +{"id": "train_040523", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "24fef6ad33b006d7ba0f8568b847ee18bcf520ec"} +{"id": "train_040524", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9dc75df125834872a15d79bed328f0f4bcd1e816"} +{"id": "train_040525", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0c7577466cbec98997a4dc2524b0856110c832ad"} +{"id": "train_040526", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f8df67c70609fc135d1b2f4e3cebcc459cb5e1ed"} +{"id": "train_040527", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "69a7735bba98017c670d75c5f9ee9d7be1de6a21"} +{"id": "train_040528", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a995bad63fc6c8d2d402a993ea3130e63cdeec49"} +{"id": "train_040529", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6d4bf5c087fed54945452df09c8b8fc56ec3dc96"} +{"id": "train_040530", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "cfacb4a57288b93b97502ff2c329bdbcd676a420"} +{"id": "train_040531", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "01379a56621359229936356f3d341bdb1a0567fe"} +{"id": "train_040532", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "12d304eef62004a677ab49dd07a0dbd936f0ff71"} +{"id": "train_040533", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a7bd5d56ae09c512ce0cb36a3d32bec065798d81"} +{"id": "train_040534", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "1507a709d1c4233dd981087325e10b0d21e4e8c1"} +{"id": "train_040535", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f53368a0032469ded99bd99b85ec80a0e022cda5"} +{"id": "train_040536", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "cf3f565fb1bb923583eaab5cdd11aafdacf746ff"} +{"id": "train_040537", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7594ebf4fc25b4778bc4d312c58ef82587d6fdb9"} +{"id": "train_040538", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2b0f9eb3635935454cfe4f6e7d0b1bc6e92b8a60"} +{"id": "train_040539", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f08406600214e9a03d1e251f7c7edab250b8b8d6"} +{"id": "train_040540", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "978e8299753f655638ab0e465225e827cf895137"} +{"id": "train_040541", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "005cf6f07123af00845a56567dab8d5f996f1f3d"} +{"id": "train_040542", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "45e13dcdcb1d2183ede6f6102d22ff1910844214"} +{"id": "train_040543", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "763cde722ac163fa764b888f0a0b360345700d70"} +{"id": "train_040544", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "690bcb9ace4af537d83a9de4c64e4e20beac58b8"} +{"id": "train_040545", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "02a56219a402b28f1540a0b0d22295aeef09a2fe"} +{"id": "train_040546", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "efa4e123044ca400bb4fd3179f4e539bbf142dc2"} +{"id": "train_040547", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "28ee71707907159e807856ab1ad668fc80bf9062"} +{"id": "train_040548", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "052e30ed22a5decff3cec73cd8502265a09a8dec"} +{"id": "train_040549", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "21ab7502d52ec404152d0c5bc5af38ebfc4fd41c"} +{"id": "train_040550", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "73c383a8146e780c614b8c5923a8e241db27c7b2"} +{"id": "train_040551", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e34c5be7a21196b1d72347c30d2ed499c31e46c0"} +{"id": "train_040552", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0d3ac99f4cfac5a8a67e094997cf9dceb7023596"} +{"id": "train_040553", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a5bb6fce36b558e31ea139f3bdb408d37411cc48"} +{"id": "train_040554", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "75403a37359dd79d988bb39b1eeaee2198b9a547"} +{"id": "train_040555", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "86edd581571f9c1cd4898db47fc1ea802d150a36"} +{"id": "train_040556", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2081053be27b5b2c6bcff63f0a55994d0e5243d9"} +{"id": "train_040557", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "41b9e632178b5b5070705a2b39d6ead97a194fe8"} +{"id": "train_040558", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "2f386dedee7d5070a112e01f22cce949f36841d8"} +{"id": "train_040559", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "184167d197ca05666bd3a8a42e2a7437fe37d1d3"} +{"id": "train_040560", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6a3eb56758e3fb20a8efb7973db04a9f463fb533"} +{"id": "train_040561", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0a8a7f385fedfbd233a59e61ee590b82e3389639"} +{"id": "train_040562", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ea5a1053cd0926b4bc226e942ec169674bc638f9"} +{"id": "train_040563", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6907e7514f933b7b5b07ada21836912cca4dbecf"} +{"id": "train_040564", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6ab65e5bb97549afbf5a629d08dc25916f9d34db"} +{"id": "train_040565", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7d036c28559e6091b60edff3a61c2156014572c9"} +{"id": "train_040566", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b4d46133b846c2125626c0d0057ee890d65233fa"} +{"id": "train_040567", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "00dbd27d1126acb24fe7742d90bcc3ac509282e4"} +{"id": "train_040568", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "227d21d22265f4224751b872d405b7b5c8c6accc"} +{"id": "train_040569", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0d712c993ccd69fd493c7e3450dc18e1027f9071"} +{"id": "train_040570", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4fa5f652e7ca4f6d60e8b73a73cf3cfd19990f5e"} +{"id": "train_040571", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "595d032ec62efd28dcd23ee7fc5446188903462f"} +{"id": "train_040572", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "6aca16f46e183b1481b9eba1e7481ca73aec8de1"} +{"id": "train_040573", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "135b9e62418640c47c34805425157c2486ba0fa8"} +{"id": "train_040574", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8d3855e332e9b440d3a161e80284ae3665468f3b"} +{"id": "train_040575", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "af5a4d1dd3cffb61f026597dd91024b7f40f9395"} +{"id": "train_040576", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "20aa2982e89f540dab68c1b068d45326ecbba9c1"} +{"id": "train_040577", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1d1f944b94bddf22d41edb62e61ccad50ff4fb0f"} +{"id": "train_040578", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1d403a31d6c5dc0b0fdbd59a1372800a03a7c502"} +{"id": "train_040579", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "59f2840b0584abdfa6f28b7447fe7fe2451e4491"} +{"id": "train_040580", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "fa0e03789772b9bdfc1951b02981252984b6bd3a"} +{"id": "train_040581", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "36d325a04986084b2b905af83fa00017e8afb9d7"} +{"id": "train_040582", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e27be762d6ff128f7c58efa7089038e130bf29e4"} +{"id": "train_040583", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a4acca85eab2641f31cfeaedffcafa20a1f722da"} +{"id": "train_040584", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3339a1ca2f6d8fe36bd7dd5032b8249230160850"} +{"id": "train_040585", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "53f1c6da26fb9e0a942cf10c45f56e52bd537db5"} +{"id": "train_040586", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "84d2f97407d444d8b42213fea999b1a4219dff03"} +{"id": "train_040587", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "32da3bc8442c39d261a892a5e88bc4aa3fa08792"} +{"id": "train_040588", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "db9f191d375e238b941ab2843856c071b7291fe2"} +{"id": "train_040589", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ff502b1f0124dcd39a42ec9ab431fff7b513cf22"} +{"id": "train_040590", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "eb149df4b525832721687ad09f30b297e1f6e462"} +{"id": "train_040591", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "086e11cedf3e4295c3113e33c17bbf9e53d40502"} +{"id": "train_040592", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3045a729efed5659097601340f0fec5a62e09278"} +{"id": "train_040593", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c6055eb15fcf6729b3ec363e3523f5224e545c15"} +{"id": "train_040594", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a40ba009b78d20e748583411f33b1ac4fd6904ae"} +{"id": "train_040595", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "6c5e8b08613667a6df2bae3e5ac98221fd61edd6"} +{"id": "train_040596", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c398013de3dfb709c6f51531f37a9e9a84f1a2e2"} +{"id": "train_040597", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "1ce215d6fddf32f07f107f38ec048154babef05a"} +{"id": "train_040598", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b1be0c5f917f6042cd8551aa310238634f3f665d"} +{"id": "train_040599", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "9e21465bcc63ecafb1166a6b5dd65a24d956a02e"} +{"id": "train_040600", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "1c7388c4977c3c72d07279c2e864eb410ee31318"} +{"id": "train_040601", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8bfb8ec502a8a94f8c4338a92c7694bb408ccbb9"} +{"id": "train_040602", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d53ee6e5fff4ed52804da1539e1d83d503ef9d6d"} +{"id": "train_040603", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "dfa4c547546fbe64ed0bbb9f960ed3688e3f3f3e"} +{"id": "train_040604", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "1b170f75f504f05ed9530d4be83701b586a11855"} +{"id": "train_040605", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c9cc07ab9f264c1463316c1650cb4b0d0742ebeb"} +{"id": "train_040606", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3b2d2e8fe2a42346615994ca88763fa35f8549f0"} +{"id": "train_040607", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "3fda35d44a6d7dbada0b6582de38088a0bd9eff7"} +{"id": "train_040608", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f24f69aa557419e0fa73b53d6894889117242db3"} +{"id": "train_040609", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "437c93cf32c47b162f1c780142f7751219c241b0"} +{"id": "train_040610", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ee28b0627b9d28a1ce4bbaf4901a57f9e304554d"} +{"id": "train_040611", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b6ac84cf5ffc7c5e9e749eee9176217f5aae09de"} +{"id": "train_040612", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8f1b09a021d998b22932cd837740de34e163cf1b"} +{"id": "train_040613", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "61fc020b2f0a24b646886a8ee4ce525e7275d23c"} +{"id": "train_040614", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fa165afebb564425a6071b2d81c5fc3b38ddf133"} +{"id": "train_040615", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c456b488aded9ef671d1704ca7880bf066b828d7"} +{"id": "train_040616", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d4ba4c39e7d907664f6d93d390c2ae07ba44b94e"} +{"id": "train_040617", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "bfe568e730b2eb492d222416c94f608133899e97"} +{"id": "train_040618", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b17db9cc24d9e84fd7f2b3edc1529388df74ecbc"} +{"id": "train_040619", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f7eafae2c5f202bac64ea29e5de96b5fa52f7b9b"} +{"id": "train_040620", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "98a086857620a3ba803b3ee512cd165b7debb8f9"} +{"id": "train_040621", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3cc9a8d107f3ca7cd057248debd3a56880e7c07c"} +{"id": "train_040622", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4e4ebb69d2646a383dcdd998db5e9717061454af"} +{"id": "train_040623", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8568ffa69749b6adaa2490970d233c80566eb883"} +{"id": "train_040624", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4fe1ff5a7a2a70f1acf5532fd3ccca4deb998eee"} +{"id": "train_040625", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "754a4fcc28976b7c5a36813010554a5f70a3d308"} +{"id": "train_040626", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "81e2b0951190ec9096db73fab973918e701547ea"} +{"id": "train_040627", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "23e0ee7d90cd7583767f398ec5ac5b8c3871bb53"} +{"id": "train_040628", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "cb2b17a49c794e17c16b4c06e5d71706ef61bb14"} +{"id": "train_040629", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "be198f0a76111d0292256128541274c9b3a8f946"} +{"id": "train_040630", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8907c41f67ec18dd49292ab809eddf0bb0739887"} +{"id": "train_040631", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5013d04ee9f95ab9be24b6b762d5e43ec777993d"} +{"id": "train_040632", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e9d57402f76d22f159ae801094d75fe3b2249aa9"} +{"id": "train_040633", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f306a29ae3ec4200397e5f038cc1d660ca852d1b"} +{"id": "train_040634", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f5575fc6818ac86d66168760f255454f5f163fcc"} +{"id": "train_040635", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "50160b93fad4399984fe49ca71f2f9509c8b01f6"} +{"id": "train_040636", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "24e353e0161516af42f593435b66851382b5e6a4"} +{"id": "train_040637", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0e02ad4af959338807dc92ca6cb620c2ba071605"} +{"id": "train_040638", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c32a5345c3b9ce5130b39e1c7e9cdc5b2b7abc48"} +{"id": "train_040639", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "93691cbf3cf750f24629f60fffec29fbff5d50e0"} +{"id": "train_040640", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "268772bd07c136944550319eaf16d90f8b429bae"} +{"id": "train_040641", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a7722fd6b2a3bcd970247eaeedd3ed400c92ef9b"} +{"id": "train_040642", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "576dd04c9e43afbe2993edc6d11f4953b5ffadb8"} +{"id": "train_040643", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5ec2e0080d780c03dd6ba1b1e4321d83a6c6b448"} +{"id": "train_040644", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "81d1be6218f7ff27a607d4cfc9ae1c0496b58b36"} +{"id": "train_040645", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2e86bdc7fa3d95ac2cd676f4e16bee7909138b94"} +{"id": "train_040646", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1a75997f6689f1bf1a8af1b5d590ba35465595d8"} +{"id": "train_040647", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c7e7511c4e6dee162725a50feb876467081cf356"} +{"id": "train_040648", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9b8ffdf715651764feb1e3d57c72101f1bd8db4c"} +{"id": "train_040649", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fa1006a21358832350ff452c33c2fe9ce554d230"} +{"id": "train_040650", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fd106ce0120241d47b7e8a8f4692081dca134851"} +{"id": "train_040651", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "250a87301e3ceb99f0c9773de6712856d8a26fa5"} +{"id": "train_040652", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fa1006a21358832350ff452c33c2fe9ce554d230"} +{"id": "train_040653", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f30cdbf1278246c6a063a43937f8328ef28abf0c"} +{"id": "train_040654", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ba9da6607d0c533fcf287c0ffa992547472f2502"} +{"id": "train_040655", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "914fe433a13c4e65c0a2747a7377c4d498711799"} +{"id": "train_040656", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9ea75eb9497f6ae5eacb3077ceb40e7514c73b0e"} +{"id": "train_040657", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "745c3671c70a5a8f281386490c84ef4e928bab1c"} +{"id": "train_040658", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fbb900d777cde3df4c6b2810fe5d7d7868b766c6"} +{"id": "train_040659", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ce131a4d9580210f0c110101e1b15aa451019f28"} +{"id": "train_040660", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2e376b493a22c1486a74ab5ab4cd1ed752642aa2"} +{"id": "train_040661", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5a88e5e67e2974523638146b7089cef56083a1be"} +{"id": "train_040662", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "17a796ad0acec145015123940e2e8d980eadb7de"} +{"id": "train_040663", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7f65171d2af83cd3bcfa703a9001c70046922908"} +{"id": "train_040664", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "be705a76348c3461ad58c9b69183f5078c6602b4"} +{"id": "train_040665", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2e95a58d5d838c36d244a2289bab01490f2e11ba"} +{"id": "train_040666", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "025ebb27ac2fd8372f0821ef75d1d100b5fc172b"} +{"id": "train_040667", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "483c8d681f946885d4d043e04420d3f058d67c0b"} +{"id": "train_040668", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8a0b9e2c6c4a1d9cfb51b2f8c008be1242625128"} +{"id": "train_040669", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2bc3cf10648a9df1fa3b36f65b502fe59161431b"} +{"id": "train_040670", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7d72fa70eb3f949d0eec909695c96a9676b940c1"} +{"id": "train_040671", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f29f745f5ea4ba29daf3d2d270d39d2a55d52ca9"} +{"id": "train_040672", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "dcf22b3b82ceb1d86a8ebe8f32b45d5f0e31bf9b"} +{"id": "train_040673", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0e41438e2fdc8cce8dff5bad1f66a0c7b8aa6056"} +{"id": "train_040674", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0eb64a143e854459490e1cdfb6c17f3c1744c2f9"} +{"id": "train_040675", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "23526f0d25af6691ac46a1088b2d3e3f721388c8"} +{"id": "train_040676", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "fe21e290dc30f92aadae321fb61cca9e3e948d16"} +{"id": "train_040677", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9859cf2f0a98f6b1a3db1f6c508b00087d421307"} +{"id": "train_040678", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fd490b9a78778bd8fa0ba15ee38c2329230371b2"} +{"id": "train_040679", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "4dc4800412318dff6bc3ac61233e4d79a8c0e490"} +{"id": "train_040680", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f76bc4a772d07d21b5e821457936f0b12a91df29"} +{"id": "train_040681", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "805b9711f10fca2cd9b8c5e38c41227bcca3cdbd"} +{"id": "train_040682", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7ebd4a59790ae9b4b6e69feea8842720c81242da"} +{"id": "train_040683", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c8cd5d52f227c30d79b711e80c4bd8ea5a9fd8cb"} +{"id": "train_040684", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "35d11af3d4d896444a088822c040eca4d3bf5b67"} +{"id": "train_040685", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "db7b211b8eff1c9fdc6a38473dd6ddcf268df248"} +{"id": "train_040686", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "791e7f72f39fddee7bd68a28223ecb15cf25cf79"} +{"id": "train_040687", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2aad973126e5a000e2ba0a7844228b17d6d03f7d"} +{"id": "train_040688", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1220b7b72475d6ece888282b3bed0510fbaae575"} +{"id": "train_040689", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "cf73d2de04d107b2b36a3796d5dc03f6040c07a7"} +{"id": "train_040690", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "00e5bf74bdea7a20c9fe10e6926094699bd47bf1"} +{"id": "train_040691", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "594b2e35fbc49b6b223ed1807cd0c8f35513b64a"} +{"id": "train_040692", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "15e749214bd610bd5c8443d1f5b8054e3ce291f0"} +{"id": "train_040693", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "dcd895150cda878fdfc6b63d43a84fccfe35eb1b"} +{"id": "train_040694", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "da148176109d6567afab1cc258a4cd71fdb54eac"} +{"id": "train_040695", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1ce29d25fbe68c3a54d9176722159ad593d51367"} +{"id": "train_040696", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a2ce7c08882776fd71aeeab96449e271b4ea9139"} +{"id": "train_040697", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a05fcedcc8f4f4cba36dd37d4a1919aa92e34ff4"} +{"id": "train_040698", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1008e16d5c2a89aa6e8468cdec2f3cb888356c79"} +{"id": "train_040699", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "44837270b443e2017473c0e275d405f10c3750f0"} +{"id": "train_040700", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f44bba0cb3b8e735794ae260e84f9d73db20eba6"} +{"id": "train_040701", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d0bf2bb36419548ea6b57a45cf5e4ef146b8d7f4"} +{"id": "train_040702", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3f3f92162e1f786bf4743186d80fa9527f0856ee"} +{"id": "train_040703", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e302b1fe904a059294e575177aec11fadc3d65ea"} +{"id": "train_040704", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c5a0ad9691d15c4807e90ea329c43ccdb2f8d14c"} +{"id": "train_040705", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "bdce22dc73ecc87e8c8b2c810c183a08112be242"} +{"id": "train_040706", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3959520789d8dee89caee9c042938dfc015f96fe"} +{"id": "train_040707", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "29e07740cde3c0786cb2c3db4212b32f1b0a84dc"} +{"id": "train_040708", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "76f8b6577a36eca868c2edbcf145b3533b5cca61"} +{"id": "train_040709", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0f59a3e94f0602be1f801891c64904cb3f0b1c65"} +{"id": "train_040710", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e57790df1a6880b33f078ebfbe3b0e296c2423f7"} +{"id": "train_040711", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "11c4278893cfac78edc4eca69ad21425c69be431"} +{"id": "train_040712", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "874fd218fd3d5dd9b95bd30bb3ac6922f2241393"} +{"id": "train_040713", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8ec7a4fa68a64ba9c65aaaa82a4d399f078c572d"} +{"id": "train_040714", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3cc058dc50271eb2fa8786ebbe9ce20f0118883e"} +{"id": "train_040715", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "94a80f854359d3e8e81848eeee486133f5a794f4"} +{"id": "train_040716", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9e17c39d5e8d00f8afafc770d1029d0dcb123f64"} +{"id": "train_040717", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "acae3e828a78da1aea95049579de909d907f4094"} +{"id": "train_040718", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0bb10d23c22daa7b5b300ff34c94c8b583187344"} +{"id": "train_040719", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ceac9490b66c1623e33cce497d89f9e5661e3ce7"} +{"id": "train_040720", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4b04cf69bf977704c75f440590419d644cdb87ba"} +{"id": "train_040721", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ea596c6354ab0088811211fbe6afa1e1a0d8b31e"} +{"id": "train_040722", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8433c91b98bef5db47c2d67cc5dbe23aa42e404a"} +{"id": "train_040723", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "aa12c88a2d5535d392f5bb246d5d7ed1a0f092df"} +{"id": "train_040724", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3e409f039b3fb158be4413da98292b3536ef5b10"} +{"id": "train_040725", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c932fd7a65f0f0fb2aa81d2d434dd5469a3fa8ff"} +{"id": "train_040726", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "221a70a0c8351fce1231d42162bbd1707ca8dc03"} +{"id": "train_040727", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "18142e92270d5d9f84c82975ac68daed151c01a4"} +{"id": "train_040728", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a9fee9abd5ea1e1ec82dd11b503b271c66e8a5e0"} +{"id": "train_040729", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ecdf8ed271be6b412c21fe053122471db8ff5c32"} +{"id": "train_040730", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "78c7202ac852362798c14030a0131420f134580a"} +{"id": "train_040731", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "284673fd58ed41287202691d1a68113ddbe28131"} +{"id": "train_040732", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3d620b6bb95a54a08071bc97a398f7ce40a084ec"} +{"id": "train_040733", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ba82bc65b6d0974c547315433938325c5e086b27"} +{"id": "train_040734", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6e164f17184df5e3027b56f06d225384a4fd4198"} +{"id": "train_040735", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5ff8a55f5217e205ba33e2d57f77306407c8d0b4"} +{"id": "train_040736", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0bd6eb60727813a06ecefc257a355b3fceda64be"} +{"id": "train_040737", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7e26da5a7edfe8ace3bb98a5cb922bcb44ea6ab4"} +{"id": "train_040738", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f28e71735e730c0883a0e41f491c6442326bb766"} +{"id": "train_040739", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "db9f191d375e238b941ab2843856c071b7291fe2"} +{"id": "train_040740", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5392bb25dc5cc63b10b9233cb500c3a0c157430f"} +{"id": "train_040741", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "405a5d4865d00bb1c39af150846de162a17f5e9c"} +{"id": "train_040742", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f6cb0a30188a07967894a0b6422d55895c5e4c6b"} +{"id": "train_040743", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2c03bf47d682edc00c8b0a238a636d02e8813b07"} +{"id": "train_040744", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "c04ba525844d1a45e6f50a66b193129e64aa7e0c"} +{"id": "train_040745", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f12875779e0ac0ee302a2391fb4b0ed547480e10"} +{"id": "train_040746", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8ad99ff4dd0a537ea43b3b99f437f03537ab3b40"} +{"id": "train_040747", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "fb3450052922af191fc49990d7559b2d103c5b46"} +{"id": "train_040748", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0e7a830edf60a600b899c208ef84b2f58d80b290"} +{"id": "train_040749", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "51ca871653d720e4c369c9bf44925e5cba6d852a"} +{"id": "train_040750", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "54687668aae90c356d36b8b5f3b6e197b756b726"} +{"id": "train_040751", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "19fd8e4936bcca7d6558aec46c0c96120e0315ab"} +{"id": "train_040752", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1aa86001e6c5c8310562148b3db2b5b396b92077"} +{"id": "train_040753", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "dcaa390f585d42072081a34a8e9398eaf2295971"} +{"id": "train_040754", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2d3f2f2e1df80787c51ea9604315afcb1775d374"} +{"id": "train_040755", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "499854534741aac12a431998923f35e7ccd54012"} +{"id": "train_040756", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d9c29f58a07b3aaabf5ccaa7ce40058d18003c7b"} +{"id": "train_040757", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "333396aae072ba7439b1709ef2bb0d39267ae927"} +{"id": "train_040758", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d0f67068af844ceb35539ad2f787f7fae0dd19cd"} +{"id": "train_040759", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ecf5379b7c01cc64c3943c9c69ddbe3e9dee4219"} +{"id": "train_040760", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "76d08805dfc19d06439e6debd321fb842ac7101c"} +{"id": "train_040761", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ded5ee4454fcedf360ba5a6167a7112fd11d6265"} +{"id": "train_040762", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d60f9e591cfdb7efe3bc38c88530e23719597240"} +{"id": "train_040763", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8f3a2d6d882cde7e2f1ba6336bd43e28db8cb967"} +{"id": "train_040764", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5516ed0ea400204216756f398212ca835c56f0db"} +{"id": "train_040765", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ac2bbfdb44af3f03c1f7a6c46515f17249eb778e"} +{"id": "train_040766", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "19cc3eff372901d30785eea18b308cec96c47d88"} +{"id": "train_040767", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6718d222a22ad691de275c8687a5feeb126bc9ea"} +{"id": "train_040768", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f0e223698f55a3d9c8d29a0c6d98d6ad52371f0a"} +{"id": "train_040769", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "c38ff50a55bcaef29cbe1d12eebb58ae477f5f1c"} +{"id": "train_040770", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4a16f3d2feb5c82b17e4c42086035ef85550b779"} +{"id": "train_040771", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b5ad848b3a467552dc6801d7ba840a5dd3ff90ae"} +{"id": "train_040772", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d649cd2826a700b3fc8f213108441411ebc62a3c"} +{"id": "train_040773", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e6850e89a2cfdca98a009d2205bf583ceb6140ab"} +{"id": "train_040774", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f5d406d1da3a90519381292efecd9325e2667888"} +{"id": "train_040775", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2932b8cc7201af118989604965a51cc78cee46e8"} +{"id": "train_040776", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "12a2cae999e41ad0f183afaef8279bfb98295126"} +{"id": "train_040777", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "4680db292d426484d1a2f87ce3e6f855a75084c3"} +{"id": "train_040778", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "96e850d35dd523a9a9854536406d3c260db14fc3"} +{"id": "train_040779", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c0dc60caf7d16084a84cbd1f3e62d9504af2faf6"} +{"id": "train_040780", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2eaaaae44c102ea9634ff54be224c696ec1348d0"} +{"id": "train_040781", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "12d55912e3b3ae6b857d279e32fad6962b4aefa3"} +{"id": "train_040782", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "cede688cfddc8e5e472cb1ebde148c4423755180"} +{"id": "train_040783", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bdf19d1820933577a151d18760d5c454b638c181"} +{"id": "train_040784", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c955f14df79c06b51b1bd03889708e10c3df0a8f"} +{"id": "train_040785", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "926ca0d356130117ba6395d8a02949220fa435e4"} +{"id": "train_040786", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "65f1ec301939f04da1ceec79ec97baedb32aa039"} +{"id": "train_040787", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b514d3b2e98e5e07715f3d2f8bc07ab64bf98a6a"} +{"id": "train_040788", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9f1a90f8b523b3e6ccb4cb9380a044bfaf35747c"} +{"id": "train_040789", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "7e35a88458c74b4031f356c19292672b17fcd9ed"} +{"id": "train_040790", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "79a902c687acb20edcd27bbf82e972cb232d460e"} +{"id": "train_040791", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ad0fbbbc69f3a7bfc32ec147a1bebb98c2343a1a"} +{"id": "train_040792", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "56cea3ca8e5cea70e4ad69a85c908b19f0076c28"} +{"id": "train_040793", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e0eebe7440afbe70d64be2912a41478962b60c9b"} +{"id": "train_040794", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "331a70c8760677ef83d48ed8eabf6b9185e6ea2b"} +{"id": "train_040795", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c3295256f367cdd8f15ba93dea199c6cd675b34f"} +{"id": "train_040796", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "99ecaf8be1f905af9f95aaaee0bd970fae9ab02a"} +{"id": "train_040797", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5fe8601c0a4234683b92700ac784b8d4d87542eb"} +{"id": "train_040798", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "dd035b1316333fbbf9fa41aa9e96b5803eae4d6e"} +{"id": "train_040799", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a3b8149f570b1c3f70fc2f844f64db88e05d2c6c"} +{"id": "train_040800", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f949b341c1c07e6f1fcd325e47de1b724b67725b"} +{"id": "train_040801", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d6a64687c2394691b4225ad7e85a474fa8e66374"} +{"id": "train_040802", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8d47e6c4e03cfbb2132981a6b743b59b8e3b8953"} +{"id": "train_040803", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a01653ad6ca0c219d6114c188b870c09f001df81"} +{"id": "train_040804", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "22e46675768b618601ba3d1698df8246b266f2ce"} +{"id": "train_040805", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "014ffec459f97e4a4cc5769c8aca9d37892bf8c0"} +{"id": "train_040806", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e30d6413a5d1e0b839ace5bca735d7772e8a50ed"} +{"id": "train_040807", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3fd46854afd8159c519159e73dcf1a8c0a313db0"} +{"id": "train_040808", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "03a846664de842c11bc577767d061582b8682f93"} +{"id": "train_040809", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ea5a1053cd0926b4bc226e942ec169674bc638f9"} +{"id": "train_040810", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9235c03610615ca7ebdde9dbc50ff81662139767"} +{"id": "train_040811", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "19b6d58601fcf352d2cab06c8b54fb25b18927bd"} +{"id": "train_040812", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6d2da016e62b792f9e808bdec69b021935228cd3"} +{"id": "train_040813", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "6b922d0b62c9b19fa410122657de5b928e8e106b"} +{"id": "train_040814", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fa2c39be828ae73c7c5a8fced6c152b281684f89"} +{"id": "train_040815", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0b9e08c5bc03319ae736a415389c10a3aa91a1cc"} +{"id": "train_040816", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "5d1765e0aa9f9d82d2c818068345730ea4c43630"} +{"id": "train_040817", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "aaaa28bfcb6343a7f820b6de26f8c408c517739d"} +{"id": "train_040818", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "abb2b5742755f0dd5261ac8ff1bd9ec4b032c0e3"} +{"id": "train_040819", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a1cd2b18be4ef6ddc79fc5003dd09e9516ae4bc3"} +{"id": "train_040820", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "06589a2c9ea8475873e5db4c2c6258b310b68087"} +{"id": "train_040821", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c3868ccdff7bd3af171a3ee24df60aabd2d346a6"} +{"id": "train_040822", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f4df679c7ba202de10ae072a05117b8d2f57cb8c"} +{"id": "train_040823", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "604e392eea495aa3feeb3aed2436fb588109380f"} +{"id": "train_040824", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "00660d5b4761c09a2f8c2f0ff898a580a2ed1181"} +{"id": "train_040825", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "99c8fd9145b0dcd89a2e33200ac9d5da98c26528"} +{"id": "train_040826", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "738135629da7504a5efa0d822c4d63c01130bb1a"} +{"id": "train_040827", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "daafa227f93f8888c4ddcda4927706e42a56ab73"} +{"id": "train_040828", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6c3f18a6060e3b4f98f425bab98ccecc1679c491"} +{"id": "train_040829", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2ed4ca1dcf9b4b7b07c908505b47d46d5008a6fa"} +{"id": "train_040830", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "609ac42ae6ed03508e82da9bbe417359e57383c5"} +{"id": "train_040831", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1b1aade1b86243581f071e966f64f9726fe14367"} +{"id": "train_040832", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "efc5656f247c0c828fd48d9d65243c0b8f1fade2"} +{"id": "train_040833", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c275910d8dc22adf762ab950cf37ed1d90cd6a3a"} +{"id": "train_040834", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a479df8d0f6888a05e07afc639e0cea241505a32"} +{"id": "train_040835", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0f1f10d7aed1cfeb05050e93b4833eda5557e297"} +{"id": "train_040836", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "96b65c9fb066f1eb7085c40765d0135d6690edcb"} +{"id": "train_040837", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f2ce35a24e1774c0de7863bc09fec714c7f2e51c"} +{"id": "train_040838", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "47dbfcd8f185dcf74d55cd5af473b5d442b63a67"} +{"id": "train_040839", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "01a9c279a14844dc94325ec9e2c43e890441dfa3"} +{"id": "train_040840", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7cb3cd52b5b602457ce5bcc1a60f659ccf4302d9"} +{"id": "train_040841", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a71e86bd47de2cb59299052465abeeade7907052"} +{"id": "train_040842", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e648674e65be9fdd02b9e587406544d106202415"} +{"id": "train_040843", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2f60fd62d38e7e9da1770770afbaa4f0f314d1fa"} +{"id": "train_040844", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6abdb25ba8a0375b063cd3306780fc267855bc65"} +{"id": "train_040845", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c9c3777b41400cf16c416f952b36f1cfbcf37302"} +{"id": "train_040846", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a12e7e9a34c0538f1ba47bb31419b1ef2ef6a2d1"} +{"id": "train_040847", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4ff37dac9b9c1c966a9a959cc41676fce61158b9"} +{"id": "train_040848", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8daa8fd80577ddadaeeb95a512cc6314bdff9dc0"} +{"id": "train_040849", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "5ad98cacf7678b3ebdbd18e8c55ea6a5f211e550"} +{"id": "train_040850", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ecfcc0d6308dfe93b33e87469985d20c79104782"} +{"id": "train_040851", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "be4ee3c1103e3446ebdd35ce8e6f629274349d16"} +{"id": "train_040852", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f6d24548d6f1aa55c5165685cba728c3807e0fb3"} +{"id": "train_040853", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "fca5df3eeb85b48025635f9d8e58154a4ea95654"} +{"id": "train_040854", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4b75b7e76c6f21dac282e6fff409e433463c38a3"} +{"id": "train_040855", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "bd641840bbe78b33ffa920e67b1da896a5ca1a9f"} +{"id": "train_040856", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "4d0f36b39a533bf320a3775143da12f05ddb8794"} +{"id": "train_040857", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "1a39e70b7337dbfbec8032ad11f24fe98560d773"} +{"id": "train_040858", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c935343d1c7cd459693183fae8184a20dccead33"} +{"id": "train_040859", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0e735b74e01b693696dd34327a5163edea16d56e"} +{"id": "train_040860", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "70b0945c8a9af31f06326c6fa488cba399f9b51e"} +{"id": "train_040861", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f7dd561e2b859f31e100c508b1ad3ca918d8ab22"} +{"id": "train_040862", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "698804ee6110287cfdff4bee04787388b80ad193"} +{"id": "train_040863", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "447755180261688e39a205f46dbd789b81d54e13"} +{"id": "train_040864", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b43a11148e936fd7d0bf13d5303ad9493818186c"} +{"id": "train_040865", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "32978cfb55625e4466c3d1fb206c2b6cc449f4cf"} +{"id": "train_040866", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e2ef2254d0de76502a64ee8f2a3d4c6bd3ea73be"} +{"id": "train_040867", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "fc31c3ccb5398cb313384a760f6f5475d15b24fc"} +{"id": "train_040868", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a925d59d5c78870d1ebaef4f25a4b1d4ea68d51a"} +{"id": "train_040869", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2c21b3902b1fddb2ab698fae8e14b585cc989630"} +{"id": "train_040870", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "33e43ef6dccbaec29d55b7330986a2117024466d"} +{"id": "train_040871", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4169c8f23535e35bbd9a5960a51abc0be8493aab"} +{"id": "train_040872", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "da24c9cc83b6987dfa9583679c63b8652852a81f"} +{"id": "train_040873", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f68b3cff0afe292d404d38e4e0dcd2e35afe04e0"} +{"id": "train_040874", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e6badc166a8c31d68aa9aae2bda6e39f1f71a1be"} +{"id": "train_040875", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "fa55e1eaef57efd57186164227c90db08425e120"} +{"id": "train_040876", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "9fa8d32500e1a8d5d12cfe21a5ece5b17d2624fa"} +{"id": "train_040877", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "63573ee533338c315456f985ffbf10d22e38be2b"} +{"id": "train_040878", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "870e811b2102d751f8742ac542da0d7aabf4a66d"} +{"id": "train_040879", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8c582d5f1c0c710a5ab5a301f787adeb396a7755"} +{"id": "train_040880", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fccf2bd60b363641a2d6fa8e5568c6e830543743"} +{"id": "train_040881", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f5447dc603590aadf6afed12be6a18d4e470e4df"} +{"id": "train_040882", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "088fd0f158b6df203ab56ccd3f93f31a3a3ec51a"} +{"id": "train_040883", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "23694eb02937c77ec3066a552a4a2976ca0e6f35"} +{"id": "train_040884", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c240227ee6174a7f21d4d4a21e42f058d8144e5c"} +{"id": "train_040885", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6baeaa0725de6ec29b70fc4674a5e191c908a60e"} +{"id": "train_040886", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e2201fc88abf1ba41c99bef93d5f2b105fecebc3"} +{"id": "train_040887", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e402ae312225e19305f2db4de4d4a302d34b126c"} +{"id": "train_040888", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6719b528e5df6966829edc5624295ea0fc9435a1"} +{"id": "train_040889", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5cfe8993db61dae9b824d2ba30e893c73a89df57"} +{"id": "train_040890", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "54f578a4daa898aa4bab5615d33ca23bd84e27c0"} +{"id": "train_040891", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "1fe39dfc1492c7903742c4fcc99b2d219901e681"} +{"id": "train_040892", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "fd1cc7dca40914c7c33140ba3552172d556d5d68"} +{"id": "train_040893", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e4b5f4ba831f22caa16135baa79354ed14693276"} +{"id": "train_040894", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a6c27cf9786fb752f3bc674682a1ad37fb9eeffa"} +{"id": "train_040895", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "22378eacaac417211a2e802adaecd4399e5a24e5"} +{"id": "train_040896", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c20dea5f13661d0e8690f5a668867b9a09813ac8"} +{"id": "train_040897", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "bf6332474fc0ab1249d3586f8fe5cc64663493dd"} +{"id": "train_040898", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "66fae0a587584e14e6dad59a2a0abcbe4096c2cc"} +{"id": "train_040899", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f8eb1851ab5e31e3be35f4dfe95bf9f942c9c786"} +{"id": "train_040900", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "aae1c3c4929a5cd7827434c9ec22d96fbcef884f"} +{"id": "train_040901", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "c2d7c1cdd70d0d18df38f47e2a11d91eb918ad8a"} +{"id": "train_040902", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "52420326ee8c1956d606134ba01aebcc0de84769"} +{"id": "train_040903", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8cc9f845bb4d060d5b1f1b331ac357c107caddeb"} +{"id": "train_040904", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8afb61fb764949273440b06bab8c114eb190c68c"} +{"id": "train_040905", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d888a79cca5c25aa40dc5bf20d010522e07d1a41"} +{"id": "train_040906", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f28086b9a44e20332577ca2f49f2c29823c45381"} +{"id": "train_040907", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "aa701258df97ec7ace0f28e8bc5e049df79a4369"} +{"id": "train_040908", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c932a1dc0db141723da73785d0fa4ae08c1400af"} +{"id": "train_040909", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e02d44a31bbc9b36b463f571bb25ae39c5127f9f"} +{"id": "train_040910", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3849969f940c0fa51567f217d291c752549df16c"} +{"id": "train_040911", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "6e41ae3fee011ed7d1bf024742a98fdb1164b577"} +{"id": "train_040912", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4d90c6310d401b5e6a3720c38ae22cda15d4db94"} +{"id": "train_040913", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "6b8c6efe6936be64111a63deae6a0b664c67cd75"} +{"id": "train_040914", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0d327cfa51a5b2c61da65cf162e6e2ed40f5a635"} +{"id": "train_040915", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cf6b5e5fd351024d300e3aadd1b7df43230f0065"} +{"id": "train_040916", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e927d7a23de86170a7b23b38114e04ab50d8a988"} +{"id": "train_040917", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "28a9c65649de84997a60f2a34a2829096105086d"} +{"id": "train_040918", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "6e43e4e8f086f211a2530e519bbff016075a4370"} +{"id": "train_040919", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "771aa04e9c4b3371a824700c673e227279bf2f28"} +{"id": "train_040920", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fab4987f5beffcc2d20326415c4c6367fabafa1e"} +{"id": "train_040921", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5de9ef2b46d8c2eacc2ebac629d94ae7f8e21004"} +{"id": "train_040922", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "028bbfd8a485a0d3b905582ad7724966a0cd91e3"} +{"id": "train_040923", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "89544422e432509e87e14cb994c65cb09e41e5fb"} +{"id": "train_040924", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "be6382eadea95bab6807f4b0c6e833afe4ead607"} +{"id": "train_040925", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b109f625d86d250cc020e3c88c2a6447a750ce79"} +{"id": "train_040926", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0428655c1464015e8185e714dbca4c033e59f43c"} +{"id": "train_040927", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "384cb83d6350a86fde6a5eff93d85dff8387f76d"} +{"id": "train_040928", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "51ca871653d720e4c369c9bf44925e5cba6d852a"} +{"id": "train_040929", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "93e87cd216e5005cc3d4a0f6e82bd821a6217338"} +{"id": "train_040930", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "45d8975457e32eac9dcc9876fc6931341112ef98"} +{"id": "train_040931", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6b42a98f8c62d49aa6343f605305598039d3152f"} +{"id": "train_040932", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "bb88e7b7d2712593e75b3f3c58d553da2032b8f7"} +{"id": "train_040933", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6b0c8d0c55b2d9a2ae6c9409052ee243a3c92799"} +{"id": "train_040934", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "751c0f3d0d531d570e9429c3c0881f858174244a"} +{"id": "train_040935", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9a77b8767ffc709f9fed810f689387242dbc8126"} +{"id": "train_040936", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "710479b385359edd4a1140097b03cb115ba77551"} +{"id": "train_040937", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "82e26b07cc04b0c2f67bbcf486a2bcd477bd90f7"} +{"id": "train_040938", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "e0980b656848d6d02c2610f1c0d214d5cef7e625"} +{"id": "train_040939", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "30fd79be8672564062c881212c0d77183c8fedce"} +{"id": "train_040940", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "88a480b363c62279c91530e52f477c493d9d66b1"} +{"id": "train_040941", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "22cf7f7f6645479f3ce675b67dbbf0672a562621"} +{"id": "train_040942", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "03128225570ef26b8a5b54745081c2de0ba77bed"} +{"id": "train_040943", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5f9ec6f093361f4a1b5ee153be5b7c8c5200836a"} +{"id": "train_040944", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2ebac9e136a595748fc8c8852a941f0f65c694f8"} +{"id": "train_040945", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a871ca5cb56a66f63fed7525265e5de60ee9b6c9"} +{"id": "train_040946", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "48162c789d7427adfe1ee66f58b7ce7815f9c22c"} +{"id": "train_040947", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9b7b726e1b6bd89d84a5a30dfdfe120cbf759824"} +{"id": "train_040948", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "fc49fa1152085e1f4c9c4f8b2dca0dfcad593052"} +{"id": "train_040949", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e823d97d25c01f1980221e1a6fc553bbfa30f86a"} +{"id": "train_040950", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "68321ed0472f4d85353311473a6c1835d72629f2"} +{"id": "train_040951", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b09dbfadea18516cf79f76d345b9a2b0937d003b"} +{"id": "train_040952", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e7ea49b24165988af212cb486aeb7dc6f1d5e635"} +{"id": "train_040953", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c1de3104cb9fa993b41e1fc5719587ea53eb409c"} +{"id": "train_040954", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "13ff421d450fad48a9b0c633c75317c56ab9e73d"} +{"id": "train_040955", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "271c641dec6e1854f92510e5abca7a48bdf1dc26"} +{"id": "train_040956", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "943aaba7a371c1fdb692fb087bde346f888145ac"} +{"id": "train_040957", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "00176edef004ffa2eb91253efd9adde9fa7893f2"} +{"id": "train_040958", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "dcbb4246032b474761d8961c44f8e89aba8ff304"} +{"id": "train_040959", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "aca8c7c939727b7187241712b52853ca215b1268"} +{"id": "train_040960", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "20e09f2d4b0c4664043dad12951500eaeec8198c"} +{"id": "train_040961", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "00212f0aede1929407bbe314a48855fd09c0a447"} +{"id": "train_040962", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a11361e816d9bb09f1016c9ec06cf62e438e5716"} +{"id": "train_040963", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "67428af2cd1f9561af2ee209ff2435a641b8539d"} +{"id": "train_040964", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7adf01949fdb8a508e41dc7c13f8abc9abd9b6ac"} +{"id": "train_040965", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "547fcff9fc4c4cacb79498987f9d3a77bc570749"} +{"id": "train_040966", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "dd13de73b164d2422542540f669751027a17c3fc"} +{"id": "train_040967", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a599a56194b7a6f0b9e9c325911af84673d406b8"} +{"id": "train_040968", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "3e98f877502163e3b302c1df9d581028e34e5f1c"} +{"id": "train_040969", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "fd0e0ab67c71842640000f048dbaa61e90b624b3"} +{"id": "train_040970", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e01a9c2fa42c027bc6b886a1088e7c2344586a78"} +{"id": "train_040971", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a590c8128393f198c5a0c7435d67951d692c8bb8"} +{"id": "train_040972", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a4acca85eab2641f31cfeaedffcafa20a1f722da"} +{"id": "train_040973", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "c82b51251c29c7ccca20327eeb7244da38d46dc2"} +{"id": "train_040974", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "df8991d850f9093c9b00323c892d02c050d33673"} +{"id": "train_040975", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f692890e518b886f54c56519cb72f0a2e637456c"} +{"id": "train_040976", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0064c6dc43c673d7d61bb6eb6a058bd862769539"} +{"id": "train_040977", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6f620d6a7c2307d6823780e83df700f49ed8d5a5"} +{"id": "train_040978", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6f011b2e04fbf6281040b09b4cd3af5614bf77c3"} +{"id": "train_040979", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "461ecb93f25ea137385fa5765a2ed906d69c23cf"} +{"id": "train_040980", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "4cfd282fec47cdfd4cc65c5973aa73b23249c177"} +{"id": "train_040981", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "352144b6d2b503511b57ea782cd04f1b0068af1d"} +{"id": "train_040982", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e3c7cdec4b13a33b09b2b9778c5c02ebf77ef098"} +{"id": "train_040983", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "22701467f9e4fd979cceef0484a0046e75ac80d5"} +{"id": "train_040984", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "fcef62755b94b13d23595b65e24e70eda0457dd3"} +{"id": "train_040985", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3ae2fc8d581c5b7437435e58aced22bed1db562c"} +{"id": "train_040986", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9b104a63e37b326c75315c911c319d57c53ba905"} +{"id": "train_040987", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fdb11aa9c831c893ff4050493d91404023500126"} +{"id": "train_040988", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "63ddab33410cd39fc2df9d71fbfbd483bb8a452c"} +{"id": "train_040989", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "af800bcc9d57cefc93c725f95d44571b741a9105"} +{"id": "train_040990", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3c681c0a5743a7257bc1fd951cbdc780b05c5f3b"} +{"id": "train_040991", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "382d1a9cfb7dee56f61416c3a4f67c6a6a484bee"} +{"id": "train_040992", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "51d05976f455fa7d085a6d91bc5d1f8ffce14408"} +{"id": "train_040993", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "82554f63b9976e48f5728c985d645325306d5eb9"} +{"id": "train_040994", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ed72ae003db6d1203c9757fd56860031a4741a80"} +{"id": "train_040995", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "182edccc5819e831d281a8b6048456a8f570e308"} +{"id": "train_040996", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8a6bdef1422a98cac26e4a84bb5e3f9a232e202b"} +{"id": "train_040997", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "03dacd18b0fdbfef7fd1c0a6799204f0d6fab1c3"} +{"id": "train_040998", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2557a7cb433a17c95de081166c2aa08e493cefb4"} +{"id": "train_040999", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fd499a704f15acaae9798eb819997a4746fb0261"} +{"id": "train_041000", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5e178342408bd8f2dc9fe423d62f1304bce47245"} +{"id": "train_041001", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a5f466125aff5f3f85bf85fddd2241215963e79c"} +{"id": "train_041002", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "89437a41f0f85472227932455a693d8a162d2c13"} +{"id": "train_041003", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "edb4641cce223088a723d96b65c4a81fa37244d8"} +{"id": "train_041004", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "017d58f5b7d5650f655aed2998c2d66ffceb48ca"} +{"id": "train_041005", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "1feda04145ea2a14cd431293f6b8f333c6a89b78"} +{"id": "train_041006", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7bbaea093756aa55e59dfc0a2c933c357236efa5"} +{"id": "train_041007", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2bdbd87ee419f3ac93de4c69b9ad7f44f35b3d65"} +{"id": "train_041008", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f6aaac1a30fb31613ef7996470a048b6d18b0608"} +{"id": "train_041009", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d7ca8155eb72d41d7178783da791f0c96588ad18"} +{"id": "train_041010", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8fb0a33b3879fcc4d65c562a5f7a8890e85e7092"} +{"id": "train_041011", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "33335c31c113dec10913d9b94dcbd48c122385cf"} +{"id": "train_041012", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "86b54121030b3c09e9178ffe14808619db8496c9"} +{"id": "train_041013", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "46bb19a840d62fbc35f3c39b692778b3d39aa73a"} +{"id": "train_041014", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "52377a5ef0d0849d19c8c40a0cbd9d82eceae77f"} +{"id": "train_041015", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6c48ee0c96dcf577abb2cad46d2940af3a61c7a0"} +{"id": "train_041016", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "90919dd52e99a2a2b91703ffe382fdc5985fce21"} +{"id": "train_041017", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6ac83ecaa6a787f054e5dc95a7c7a62b1a6f2dce"} +{"id": "train_041018", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9ac78873b5030e71c0cd58585c9b2a57fd423e90"} +{"id": "train_041019", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "22568f77bd04a421ec0433bcbccd89499ac215a2"} +{"id": "train_041020", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "31c82ffb73e5baeb656bd3d763b754b6d004bc3b"} +{"id": "train_041021", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a4cbc0060a589adaec3c5201c3377990ba2dfbaf"} +{"id": "train_041022", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8d73f03ec9d9531ffd8032ace198e6e5f2d22617"} +{"id": "train_041023", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b757ce205a3479d442714aef8d6cb33315013a4f"} +{"id": "train_041024", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f5d5fca7d9335e2867172619b4a07969ab3e7954"} +{"id": "train_041025", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e048f6b0343bfbb24093da390e100781e38892ff"} +{"id": "train_041026", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "9f337afc8a6a63e96fb605bf59ea6d90f13bd50b"} +{"id": "train_041027", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "10822aedeab2b5cc325c580e75b58691817e658d"} +{"id": "train_041028", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "809cdf85d6fe167ebdb99e1bd9b0fccff576a5a5"} +{"id": "train_041029", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "68a00b97c7237956d491996506c031a752f3f5c5"} +{"id": "train_041030", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3ea23bd34645e39dbbc2fce46a03eb1c18b799cd"} +{"id": "train_041031", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8a1c3bbb01aaeb480826d2ec714d64345bfd9255"} +{"id": "train_041032", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5466df7feb7e426f2cb6752da1bb7b208eacd174"} +{"id": "train_041033", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "092485e0c3145d8a3b61101511b6c68c20f99848"} +{"id": "train_041034", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b7e9f03c3e158b89584864f55c772b5598252952"} +{"id": "train_041035", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "65d996ce87d918309595e6aeca3ce88a3750ca98"} +{"id": "train_041036", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "db57c8ddc526f7326d8ddfa9f2cff9d995ba6e2a"} +{"id": "train_041037", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "8555100dd057ef8c06d715b3b437f4d5ae6f116e"} +{"id": "train_041038", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2f3f97958568ee58b245dbe80e7a5a60d8734347"} +{"id": "train_041039", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ddd759fcd5207081ffc4f05798c17b54ddfd5e0d"} +{"id": "train_041040", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2653e5f69c1c4a17203f2845be0d551e000a10e3"} +{"id": "train_041041", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0e41438e2fdc8cce8dff5bad1f66a0c7b8aa6056"} +{"id": "train_041042", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "b1ac1fc910a3b4f3a98722b55214550c16d42710"} +{"id": "train_041043", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4de97e22bb56ba131756845c4a60bc517f23e3c3"} +{"id": "train_041044", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "cc41f2b517d3d56b5470e56076c00e9cb6ca9ce2"} +{"id": "train_041045", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d7c4a92adedb36e3c3e8b1900482a99ea95d7385"} +{"id": "train_041046", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3d6ed9289b17a7b76b0c049327043070eb031881"} +{"id": "train_041047", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ed528c9bce17f3b408b4552b246d2efd5a112770"} +{"id": "train_041048", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ae259190a5cee6cb7c0be08fb8a5b6378268681f"} +{"id": "train_041049", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b25ce8833636bd9f62fb1d17b68576297ecbce46"} +{"id": "train_041050", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "18a529c7f18909f2bdf36158d301ddd7a2e33616"} +{"id": "train_041051", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "85ecc5c902fef2b3d6844a172e160e810f207610"} +{"id": "train_041052", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9ec3b7835d76d1e361a5f00b5940eb2d8d692bbd"} +{"id": "train_041053", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "188a396f3d4be5e08d7b4b68263ab139bf6068e6"} +{"id": "train_041054", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d21734621c76bd3a72fb2a21fe654512b3d01397"} +{"id": "train_041055", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "929496d45462bb0c172f157a40a35434176ee8ae"} +{"id": "train_041056", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "4dc7cb8da1a0b3157003faefb4dafb87ed689651"} +{"id": "train_041057", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "42a49b1a080770587de92d57312d896996401a5d"} +{"id": "train_041058", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ee47fb252944a86e6fac0ffcb441b4ae2027cffc"} +{"id": "train_041059", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "bdce22dc73ecc87e8c8b2c810c183a08112be242"} +{"id": "train_041060", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9417a9870db621801c7162c78b849d2df5830fc6"} +{"id": "train_041061", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "1c227a9c9e04733deeac3be6e6ce7f666ba1cf29"} +{"id": "train_041062", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d9cc9584947052604b76ea47cf862b94752e1adf"} +{"id": "train_041063", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5ae73727785a84055bbb12ddc2f7fbc52b196ebc"} +{"id": "train_041064", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7712c0949c6358dbec49e8284300684d541b73ff"} +{"id": "train_041065", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "772cba8a127e04fe8d2c670a9d6a716b70971ca0"} +{"id": "train_041066", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d3cc7af2a721b874a805efed9242988965c31077"} +{"id": "train_041067", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "d418ac22e0648ceb417946cb697a6c330de904e4"} +{"id": "train_041068", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a833ef6c370ba2868014b667f04d7bda55acb56c"} +{"id": "train_041069", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "28e10dc59ced3f6500fc68438731174ef4751885"} +{"id": "train_041070", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a84b4694328591abf9342a67881e7a7bac73f686"} +{"id": "train_041071", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b17ebbeb1e8f571cdc634d3e951a2329b5e3f0cb"} +{"id": "train_041072", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "57b596882f11de9730c32b50739b4d7491a57c3f"} +{"id": "train_041073", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e70e2d7a970dc0a98f7e845fa5d70ce469cf9a39"} +{"id": "train_041074", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ed528c9bce17f3b408b4552b246d2efd5a112770"} +{"id": "train_041075", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "50f82dadf2428901072454dc2ce690e00d6a24d1"} +{"id": "train_041076", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ae533309dc1bbed3658f0c06c66902619bf0bca3"} +{"id": "train_041077", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f3a7b5074e4c7b175e406a41f1708edebe3f6454"} +{"id": "train_041078", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6bf28e4314b461a5144bbb1fd4dd9b35584b2a0c"} +{"id": "train_041079", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ffad3061c20e95198f0cf38f84b7a83f1523711c"} +{"id": "train_041080", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "57655d196b85ab47c0827617bdf66ed191ecd900"} +{"id": "train_041081", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a3cf646c768204ec3c5aded85b4707d1cb585724"} +{"id": "train_041082", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d60e80d5a7e4f718ff47a51255a08ef1273acb1f"} +{"id": "train_041083", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0593bc73c186c92188defdb6e5ace4afb62523f2"} +{"id": "train_041084", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "fc6ef9012628fffb699668c05c2e3854094a4106"} +{"id": "train_041085", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "1cc5c5e1dcf7fe8bf96ae2cc7669f8bf51c6ab88"} +{"id": "train_041086", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "cf53a6e5195b395b5909cea90e203a0ed153e520"} +{"id": "train_041087", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c3c6a4884c687d12d40690c0a9ae6bad8fc0d514"} +{"id": "train_041088", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b8baff14ec353f314251c9d931bfd0278dc4824c"} +{"id": "train_041089", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1f61b6b6c6bd59df7d35f0680432ad3b0d9e36b9"} +{"id": "train_041090", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a6e437c9d709ed20f8efa1a3e91209a0f7e1a88e"} +{"id": "train_041091", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "aec9f6411775855da5eb1fe61eb7ba9bb9d7dc45"} +{"id": "train_041092", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e3e4d1740ea617492b6efa9bbd018ee0064001f2"} +{"id": "train_041093", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "adfcbe2f3e962aa6de4ccb9e4f9e91f07123f8d0"} +{"id": "train_041094", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "72fbc3725da4ae9e1260a0be47328e80477645e6"} +{"id": "train_041095", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8eaa719891d6ac53f6553db1198035c84d3e6a0d"} +{"id": "train_041096", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "76f6b3542bad72b1ccce4e690c71912101083ed9"} +{"id": "train_041097", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9b95119ef4d0c71b740e742f66aeb3638faa2766"} +{"id": "train_041098", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "73aea9718e7243130e9b97a0b93a539df5fa4c14"} +{"id": "train_041099", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8ebe6f5cc1cd894b20b034d2fa2c53beef4b6613"} +{"id": "train_041100", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "3f2daaa9da64bb474faf682efd64efbe3447fa88"} +{"id": "train_041101", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "dbe65a960a63f0856f84af6197a54bc5622f4ca2"} +{"id": "train_041102", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "37be5190fd2657f30d1cdd6f32f187f13e55230a"} +{"id": "train_041103", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "31a7ee61eea3662a256bec3c06e74f6b97ea11d7"} +{"id": "train_041104", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "bb37b2665fde83547b489f413ad52575a453aa7a"} +{"id": "train_041105", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "66cdc52a9d3f21fc39227b1f5b5074719cc86875"} +{"id": "train_041106", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ae7f9709a7eb2d4b41de105c86ea617d8b3628af"} +{"id": "train_041107", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "75268d49aa3d075476d86c1a1a75f1ac4d8e2a1a"} +{"id": "train_041108", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "508a68a32c018907ef856929c644577a8485b631"} +{"id": "train_041109", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "646735f7dc0db4ac01ea3cc12b9c2b25a9a1b4f1"} +{"id": "train_041110", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "54302977e6f2ddfd51537a45e11964fe71d97743"} +{"id": "train_041111", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ab08ece87e295142f89b4aa1848c7e1ed84daeae"} +{"id": "train_041112", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ac6453bd6eac93869cdcf7b4d09cbe8082c168c1"} +{"id": "train_041113", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b78924a3ba0390ade31a8ba720edf9a595bb91f1"} +{"id": "train_041114", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5585eb384ace4a9499c4cc51b3bdbfd93e2d5743"} +{"id": "train_041115", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9893c1c3acc58ce9990f3bd4ab69111bedefaade"} +{"id": "train_041116", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "07f1cdd0a08c214d3b2ca123bd0764d6bd3dc3fd"} +{"id": "train_041117", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "01f0ede8641d7d45d31b57ad26694d5fea03b732"} +{"id": "train_041118", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "441dc6c7ca1c23c7df820b7a9b9c4789c9d388c0"} +{"id": "train_041119", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "191d09fe44beb77512481476dd6a32d2834771ef"} +{"id": "train_041120", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "980fdafc0e3729b5d655712fc2ac4735f58f8676"} +{"id": "train_041121", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d60f1a6f505ce59ade25baf40384c2596131a1aa"} +{"id": "train_041122", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "da9deb67d7c8176a6759b66a87a23a806d9f298b"} +{"id": "train_041123", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "774b1e6bad1979ae3134abb54f2869e7c01a1ad0"} +{"id": "train_041124", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a0109cf0c4669c5a6c26f27d9db40ca034450fa5"} +{"id": "train_041125", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e99dfe0eb017ef4fec4d6e413b5956ab1249847f"} +{"id": "train_041126", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4e17b9af519c240601a94c0e16a7f9b2526044b0"} +{"id": "train_041127", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ca5e26e305f5f592331e429b9f585d59e2a786dc"} +{"id": "train_041128", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "32cc996e22fe1938d1840163b7304c6983260166"} +{"id": "train_041129", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bfd5fb80213496923073ef4a1f0289ab63bb8326"} +{"id": "train_041130", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "016ee965bf128ab58386bd98e14d6761f58de5e7"} +{"id": "train_041131", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f414c5f19fb0be961dcf927f2cbad7b3acead0a7"} +{"id": "train_041132", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "782096036a4a77baa025c786d1ef6301bdb54b38"} +{"id": "train_041133", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a928122813cfc5d7961e013083954adbfafbe788"} +{"id": "train_041134", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "2e95a58d5d838c36d244a2289bab01490f2e11ba"} +{"id": "train_041135", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "bd03efbe5b35c206f75a146380ce83a553313cd5"} +{"id": "train_041136", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a4027367be49d2be55a1aa78b5b97d302d7713e5"} +{"id": "train_041137", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fe91e32c56a4a31724c888f881689695ef488850"} +{"id": "train_041138", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "dffe0bbc840323f93990d3a33ceb53d8759e8a72"} +{"id": "train_041139", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "d4b7a803ec69640ea6c6723469874fc66985b0e1"} +{"id": "train_041140", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0a68e648d10fda6c25b303f77be7bad3a77e1c76"} +{"id": "train_041141", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d3f648e1a42da1bf3fea3e927d1c816c7d84b8a3"} +{"id": "train_041142", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c616e8e9114d63689eb1e500ba60c68e4d139a28"} +{"id": "train_041143", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "4cfe6777475f8b6bc0d53cbb32fa5f41a6c3c95f"} +{"id": "train_041144", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a5f8df7329d4ff36e7f99c79684ea3ecfc6400cb"} +{"id": "train_041145", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7523d93b37ece021ca8668105d2e9c11226480ce"} +{"id": "train_041146", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bf6878ce3dacf271288abd3314159f0f5096d204"} +{"id": "train_041147", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "00de19cc09770f2579c1c7ed38beca6c22750754"} +{"id": "train_041148", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4393056332caa435ac19302aa86dbf67109d3729"} +{"id": "train_041149", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "07a0965f322d79ac1ad7a64b956f94dbcfb0b0e0"} +{"id": "train_041150", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "37f971bfd486f0802b9595fca7bd105979c4c66e"} +{"id": "train_041151", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "90e64ebec8e27849dc0876854f2d87827899bd88"} +{"id": "train_041152", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5aa9ab865ba54b52c68de2b63975146e24700a6d"} +{"id": "train_041153", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "964c2c3bc25a751327c15380010a1303a7834ad5"} +{"id": "train_041154", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "fe91e32c56a4a31724c888f881689695ef488850"} +{"id": "train_041155", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8f81df1042f204bb467dc8a6314ebb8d18500cf1"} +{"id": "train_041156", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e0b6e341739b7020fff0365c5aa19661e808aa56"} +{"id": "train_041157", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "7e97f658799a72a35b3c60de1a47b3de95034df7"} +{"id": "train_041158", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a57ec6cca6327a651182a4c967d6adabcbdfa841"} +{"id": "train_041159", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3a018f89b0aa897ee34b07ad7c299bae84d9feed"} +{"id": "train_041160", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "513c6f062141cdea5b96b59821bc029f63862886"} +{"id": "train_041161", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f0b65fee1a80c0d34b2ab0ed1c3d1c582fc98cfc"} +{"id": "train_041162", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "1096bceaad3c5046d05d05983c5f7dc686b2be3c"} +{"id": "train_041163", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5d69b3865e039704c5491ee9a4680bdd73da8295"} +{"id": "train_041164", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "b58a2cb496e0b4aaf1072a76baf60c3dae81c2fe"} +{"id": "train_041165", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "20804c9150d2ba783c4f854e2666e209664cc5c5"} +{"id": "train_041166", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b88e27a3d2e44a7bd7d97e381f438687578dbef7"} +{"id": "train_041167", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7ea14f6ee0567321f547bf0813c3ccad7349c8d2"} +{"id": "train_041168", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "91f59eb0252b388715e1247d23a9acab4fa3d89c"} +{"id": "train_041169", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "bbedb87efaaad718f73a4b234fa79165b0d4c422"} +{"id": "train_041170", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f518cfff66c62b87ca1370aaa1944b08009fa007"} +{"id": "train_041171", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "245d0ff7c5ee8444d6e01fc6eaf3dec257f7536c"} +{"id": "train_041172", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "73d5b5f5f9d388a2f7e59f440d5fb255cb576d02"} +{"id": "train_041173", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9ecda924c6d2cb9c6591408fe8f78a96bfbb6b04"} +{"id": "train_041174", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8c86616364559ae4741be0e720e3dcf00ac40cc1"} +{"id": "train_041175", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "60647cf4a29a0d8438b6d9ef47acbf32915fc2a8"} +{"id": "train_041176", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1a18e489e761f82ad2a562dc25d99fae80cfa961"} +{"id": "train_041177", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "714d6cf8b17f67ec52563929d94a01bdc3f1adc7"} +{"id": "train_041178", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a4eb859ab7f853e1b2a7afdb6f78718c9ba2af10"} +{"id": "train_041179", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "cef7cb7fb3228717f9b1e1c4848e6d7e7a0f544b"} +{"id": "train_041180", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ea5a1053cd0926b4bc226e942ec169674bc638f9"} +{"id": "train_041181", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1316cf7fe2ef7ed8ae1d5d93f1a3e9038d9e85b1"} +{"id": "train_041182", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1ba1cb6a0a0ab341833c2f503716c42ae9e85c35"} +{"id": "train_041183", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "53501ca9af9395c99028258f5a3fa6553e1acc37"} +{"id": "train_041184", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "61edd534c79b209f64c3c6d96a44814b1eb09140"} +{"id": "train_041185", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5e03356b68cbb94eb2c5c9d6bb87ea4fe081e802"} +{"id": "train_041186", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b30589b9ea6465e65029c01c0c572235815c9832"} +{"id": "train_041187", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8822383f2c2ac0c8cc0851461483ab2a72a732e0"} +{"id": "train_041188", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "9399a4c3e11d84fd64544841413a20505644f5ae"} +{"id": "train_041189", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3b2d2e8fe2a42346615994ca88763fa35f8549f0"} +{"id": "train_041190", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8fd0295b9648940ea6b1a76d4fb2d4f27c996a99"} +{"id": "train_041191", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "7f7c09e70b410f221efb32aaed403677dff3a972"} +{"id": "train_041192", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "70dbfb81a07a9bb8cdb956f182bafa17a0765b62"} +{"id": "train_041193", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "05a298d01cd6dba662a277e89f0c45183c825853"} +{"id": "train_041194", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f0cebc937c8cf796d7820bf8becd56bc646eeedb"} +{"id": "train_041195", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7fdc534b4da63b78a2a44195aefa3ee0cf76a1d9"} +{"id": "train_041196", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "43adaffbc77603b53707c0f05a0f4a53da3b61c2"} +{"id": "train_041197", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "63fb30be08631884afd0d736b3efb59ba177c9ef"} +{"id": "train_041198", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b30c1e70a0fe1c4c54c4bb8178164249bf83df8d"} +{"id": "train_041199", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "cd2f0bde2cfe9a3c01ebcc0cfe4ff3387b3cfd7d"} +{"id": "train_041200", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6abd4c26981ff504675a568820fa73b51019c707"} +{"id": "train_041201", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e3760cc363be39415e22d2b593a2a79deedd91d8"} +{"id": "train_041202", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "bad8286510b7240cc198b05e8cdcc96e5d52fd30"} +{"id": "train_041203", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3247d3499a743878191221f1b07eac265618f498"} +{"id": "train_041204", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "14fc7f49450205374124727ad3906637b6892dfd"} +{"id": "train_041205", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9a1d7af713146b14f7088f8e9bf3d762f8ff79a0"} +{"id": "train_041206", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c23d55930d6963f172752017bd52edca7c7c0bad"} +{"id": "train_041207", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "44500a241d01ed4b386f2ff8948d7f670f36cf66"} +{"id": "train_041208", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c7121a68b006df21c00137bbecaa571e32201d1b"} +{"id": "train_041209", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7e079f9ef6decfd2f202f33379adb708377fa927"} +{"id": "train_041210", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6a0f4b8a44c8aa0ae3b725a6543f6c09250e528a"} +{"id": "train_041211", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f342785780e1335ceeba1fc7728a3d2abf3d4405"} +{"id": "train_041212", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a07eb14b4a9fafb2b064b5f263c56a7132c7539b"} +{"id": "train_041213", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2493c0cc4689eac58c0342f70cb56c2949a54330"} +{"id": "train_041214", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "66f2e95e61b41ef8d683cc9604fb0633d4f03a97"} +{"id": "train_041215", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cfe088ca2ee312da899618b31c293c5892e449ac"} +{"id": "train_041216", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "41f69f4bb7f1a74f719efa5ec6e33725115398ce"} +{"id": "train_041217", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f6a5221d9ce41bb00653d5d99d37f4e6194e32e2"} +{"id": "train_041218", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "258f789954d6f6d9a1f60d474705eb9e22800299"} +{"id": "train_041219", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "79182d5a1c9ef3e6198a42b91e2831bb3a90d644"} +{"id": "train_041220", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ef2ae055623054c7f9ab2590337ce91a075f75ac"} +{"id": "train_041221", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bce0bb1fa04719c743b967b6ab12b073d7fd5cba"} +{"id": "train_041222", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "72491b00073f861742cd478154c154eef230e398"} +{"id": "train_041223", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "fb96f43342c8decbc32884e15daaf7e51723caf9"} +{"id": "train_041224", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c240227ee6174a7f21d4d4a21e42f058d8144e5c"} +{"id": "train_041225", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "08eef8dedeb2e60b7bbea5005af36693e5f55a38"} +{"id": "train_041226", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e4278740ccc7e20a066f3b37c82b5bc4085826b1"} +{"id": "train_041227", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f4593d81785d45a01c1bf2db9aad9ec53d2c98be"} +{"id": "train_041228", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6b6440a740ac6f42979654eb66a7d22b3711b853"} +{"id": "train_041229", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "27db8faa4591007410c980486fba65ee34f6c56d"} +{"id": "train_041230", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "36acfbed3e33a6b57208726a795a8135d4f4821b"} +{"id": "train_041231", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "67b0377134816d7f28c3505f53e8263d2b6bcc54"} +{"id": "train_041232", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5a84e258797742bb2401ffccb7dcbb50862998b4"} +{"id": "train_041233", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "19f6abf555f095f87c55e17b0ad10dc890dd91c4"} +{"id": "train_041234", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "79021d5c633e23a03c90592a06eefde5f0f949ac"} +{"id": "train_041235", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d9242ee975f7bea792ba1ddaf4821921b1d8535b"} +{"id": "train_041236", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e9b31765ed5cf0f3cf8954f38cd8879a904fa981"} +{"id": "train_041237", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fe40a5ccc84c09bf38e0f85edc5d4ba5ac20cc7d"} +{"id": "train_041238", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "792452b9e9bcdfe476b58391b64d59d398313c09"} +{"id": "train_041239", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b056c3de31a55e804e32a9ef1d62c966f64fcba1"} +{"id": "train_041240", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "866fe132810dcc13ffebaec33602cc475708a18d"} +{"id": "train_041241", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7cb3cd52b5b602457ce5bcc1a60f659ccf4302d9"} +{"id": "train_041242", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "7cf34dda10f91500c236991bb97f66ce451d244d"} +{"id": "train_041243", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9018a58bea19bbad216caa5a7ffdf837d05ea6d1"} +{"id": "train_041244", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "88399b120fe5d35a75af52f82193878dfe08f4fa"} +{"id": "train_041245", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a6abfb8f4225a3ac57f32bbda1a802280588ad90"} +{"id": "train_041246", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "fbc203a77d29ecf2755f04703f3a9f9a4a05244c"} +{"id": "train_041247", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c547ff3494a701f38f9cb9bd3e75db9e7cb09230"} +{"id": "train_041248", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ecfcc0d6308dfe93b33e87469985d20c79104782"} +{"id": "train_041249", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "456b73c6af99d58e8422803b270bcfaaba1a9c97"} +{"id": "train_041250", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "07bcf20344b67bcf5df8c038f9089242888e0a38"} +{"id": "train_041251", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "cf0e8525fd29816f1f9c4a9c9789f6e58ca1d686"} +{"id": "train_041252", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "3d563bc639338a2d10b29cd615110ed84111d062"} +{"id": "train_041253", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "eadb4e44dd0f48daa545d5fa0b570b7b79f1362d"} +{"id": "train_041254", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ccfda8e53c6fa62e6910c8b086360bf33486e501"} +{"id": "train_041255", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7a2f3e7d5b34b933fb7b0b339c9f40c496bd347e"} +{"id": "train_041256", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8bdb9e84352a55ed7e07b2252f12698018935231"} +{"id": "train_041257", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d56a174cbe601d00f749cc81e337ce12f47a0e7d"} +{"id": "train_041258", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f8ad6e3711c251103d6e6d686297583e84eb1e49"} +{"id": "train_041259", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ef9450f1fcb77010e96aba3daa09c598f105b709"} +{"id": "train_041260", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "eb60fb3627ea00e0d73557157dbdd1a48c9ddab3"} +{"id": "train_041261", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "beaf07cd94f5b0b7419b40dcf2a541687343886a"} +{"id": "train_041262", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "85c2af5a32a94e39ca064332e87d44db6c8d11f7"} +{"id": "train_041263", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9b730f619bf08193001d22c33395505617146772"} +{"id": "train_041264", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "bfc01ad8819a636e8e4eb160a6861486f4eef9f6"} +{"id": "train_041265", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fe38c1639a08e32893002f34f6f913f3f0327b32"} +{"id": "train_041266", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0b0b2dc4b93f95483c1d22de893efcd3ebc483e6"} +{"id": "train_041267", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "dffe0bbc840323f93990d3a33ceb53d8759e8a72"} +{"id": "train_041268", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d40830cc3ab34d9256c5792164e20ac71bcfc0f8"} +{"id": "train_041269", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "753361d0fa1266758df57c3dc152b82236ffa081"} +{"id": "train_041270", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1ad37656e4e7280754148d8bf8a4ecd7a655cf5d"} +{"id": "train_041271", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0369319836a6ce66d48c21129f7f4012fd78464a"} +{"id": "train_041272", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2e0661cebe25580b2c0a39aac4db07185b1d960f"} +{"id": "train_041273", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e79a942a5bf23e7c96f3d39bc9832850bed558c2"} +{"id": "train_041274", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0f23cf7c158f6684a45a839852650346b5777577"} +{"id": "train_041275", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "45fc5e9800e3add305a60058b9e0e9b7cd5a4b27"} +{"id": "train_041276", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f7824bc3fb8d9adc44b660520c661168bf6e1d71"} +{"id": "train_041277", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e259007daaeff209655608019e1727b7ceca2de8"} +{"id": "train_041278", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "fec61aebc519c1fb9b3706b880623cd077075dbd"} +{"id": "train_041279", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0853c549da155cc1ee15670f2032ff06f3306698"} +{"id": "train_041280", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "adc71ecb3d1c724fc2c75c9504dbea357bc64d20"} +{"id": "train_041281", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f7f17b0dee8e1992fbbad53d421ceb6bb05ea70a"} +{"id": "train_041282", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a1f0fc610163ad636cd6f6f9b2d595349ff0c252"} +{"id": "train_041283", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "b282f2bd0edf18d417b8462adf96b1a79f1de87c"} +{"id": "train_041284", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a26d367c085e0c10a7cfe4dcf015282567b1ee21"} +{"id": "train_041285", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "95fc4d74b1542d1b1a4741dbbbd8c1bc9e2b578d"} +{"id": "train_041286", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "740e12552ed276d9501098bc82e9c8c302a992a3"} +{"id": "train_041287", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2898046aaa08cfca0efe2cd9c9d907c266c2b657"} +{"id": "train_041288", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "065b85a51c37d2d2cd083ca34ca8f3c287324d1f"} +{"id": "train_041289", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "61f954bca15299274332459e7b66f12b67679e7a"} +{"id": "train_041290", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "509ca49ac9fd5ff37395b32cb76dbbe8f893972e"} +{"id": "train_041291", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "406c18d0f9b22f1b50a518821c3dad51749ef2ef"} +{"id": "train_041292", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "72d2b2342a52ad32fb0aad7a4e4c8e372f1acf04"} +{"id": "train_041293", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "61f9b1aa04f7955a339c9a2a6a63cb13e8b2319d"} +{"id": "train_041294", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "670482ad4257ee4b3867e2900363359d08cdd025"} +{"id": "train_041295", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c96ba243a8852f67fef5d2d6d74d8d9b6d025f20"} +{"id": "train_041296", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d189386bfb30d744d014e58121bdea3b83869ab6"} +{"id": "train_041297", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "787dd58d6332b6ce8db1adda40d50c67f92e7fe4"} +{"id": "train_041298", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a8b5affe4d3b4a2c924e4c0c86252294a29a434d"} +{"id": "train_041299", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ffad3061c20e95198f0cf38f84b7a83f1523711c"} +{"id": "train_041300", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "31a7ee61eea3662a256bec3c06e74f6b97ea11d7"} +{"id": "train_041301", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b613d721fa27246dd6678f891b86e236f6dcb648"} +{"id": "train_041302", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f2538e47683711ca313a35ca9202848a09454b61"} +{"id": "train_041303", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "77021bcf0f8e45c76b07427ba61286073ba5bf9d"} +{"id": "train_041304", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "63a75a5f2dfcd66f0f673ed67a1dec07c71837f5"} +{"id": "train_041305", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "28c70a8df8a3c0376fb794ade82d6ff0541db91f"} +{"id": "train_041306", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "cf6b5e5fd351024d300e3aadd1b7df43230f0065"} +{"id": "train_041307", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7384c28528ef216ecceacb17d8d9c160014fd5c2"} +{"id": "train_041308", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0b3a9067201654a46a6eae189fa0be7622295110"} +{"id": "train_041309", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "196b47d2437b1af852d7f027f2c8448b41f1668e"} +{"id": "train_041310", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2e3354489ee22720deb7638a417931196aa32b59"} +{"id": "train_041311", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e71fc58ea984297ce47bb6b63ba07a6ddb5013a7"} +{"id": "train_041312", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9fc9101e946843ce032a268986aedeb985cd8c07"} +{"id": "train_041313", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c9314f9fafb5b2c0c2fc440534904fc273000f06"} +{"id": "train_041314", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6c134b4f3f5d52ca6165193846cf179107216af4"} +{"id": "train_041315", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2fe7114a21aac96a1d983084a743ab076a8abdda"} +{"id": "train_041316", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "41fdd2f570ec1a06817099ad429e6a8d1ee9fa13"} +{"id": "train_041317", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d8a7472560ebb995a1126d5f9210cdb908ccb42a"} +{"id": "train_041318", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7ab0a2de81fdd6363f62655faea67fbfe03583d1"} +{"id": "train_041319", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6003489d3453b4fe520a3a49a34388791013f071"} +{"id": "train_041320", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b5f4a07962dbb57b7292f3cb5d2051922f92d7fb"} +{"id": "train_041321", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "643dd91f09f8b50c9b9915f411cf3eb2928fde29"} +{"id": "train_041322", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4eb1063cd49d27ddcd8620aa8cc19eba3c560a46"} +{"id": "train_041323", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b6148328ab4fcf577c80cbd8eb1aea9b4e89ee9d"} +{"id": "train_041324", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "32a3d1e780b60f90b462b46441ad10d00918dcbb"} +{"id": "train_041325", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "22cf7f7f6645479f3ce675b67dbbf0672a562621"} +{"id": "train_041326", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "fc5c00dc8fecbb5f147ff3faf1a1d3e349d336cc"} +{"id": "train_041327", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "202edbb49648ad7bdaffbb94fbef4cf6307f118c"} +{"id": "train_041328", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "649b675c097e8c1b78bda97e5613f1626f61922a"} +{"id": "train_041329", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "4c2f1003f0e0d351b9231d58d1b8f469396f7967"} +{"id": "train_041330", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "10ac123abeb1675b5c112e98138d5e7c98596484"} +{"id": "train_041331", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5259f27968afe07c1d95d99ebcc37a234134f9ef"} +{"id": "train_041332", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d2ab8ff50a6f0bebf51a53e681933cbc9a3235c2"} +{"id": "train_041333", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3f5fc5f1c2c7c0b656010b8e97fc54cfd902a0d4"} +{"id": "train_041334", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6f33404993400106659c7d54bdf736e0ff53b034"} +{"id": "train_041335", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "e9f395021e43a7542908a4e60379d7e31b5410d0"} +{"id": "train_041336", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "bfc918da558e6e16124c23a8c14c3d87797eda46"} +{"id": "train_041337", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "147794813aae1471bb05ee50876e8783fb6d3dd8"} +{"id": "train_041338", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a7a1a21251736991bc0dfb27ab9a11f3abdbc384"} +{"id": "train_041339", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "bcababc196d0e89b30df78c60fc541fa4007a9d9"} +{"id": "train_041340", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "253a780bae2d44c4ae7c67230c1fb3bc4d047cc0"} +{"id": "train_041341", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2855c5d8ed22bf7a672cd5a7a054e2727bb188f4"} +{"id": "train_041342", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "073165894f90fac274fd41124a23db323389bad3"} +{"id": "train_041343", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2c4f25b84cdbd3b4a8664c455d9d596a8dc0a643"} +{"id": "train_041344", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b691579cf5790e4223289606cb78d12f7f6c9f65"} +{"id": "train_041345", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4a9dc9f6082fa526199e79425b8261e8c2a9c5f5"} +{"id": "train_041346", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3ad69ff46f138e988fc66c329808c12277d95cc8"} +{"id": "train_041347", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "368c27bdb40c4577f4b3ca72aec9f297294b6550"} +{"id": "train_041348", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9b3b892214eeb38c1f48872054146a5c3e2266e7"} +{"id": "train_041349", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2b5fce926d1d2978eeba691174fd42d3a40d4ad0"} +{"id": "train_041350", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a57326aae59cc870a97b432517de58d32bcb9366"} +{"id": "train_041351", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "7091f1a90242233eeb776b7ae3f8878161ff0095"} +{"id": "train_041352", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "41683dd1c0bab35a595981a3f28de4289f8b5724"} +{"id": "train_041353", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b6e595c66d6b1cb61320b6f82962e86717a552fd"} +{"id": "train_041354", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "c00902ccd7cc1d40ce51344660ed15a2aeb0ead4"} +{"id": "train_041355", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "fa1d1bbadf3c43b9fd094e3020ffd83e4c859558"} +{"id": "train_041356", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1f3d090901980ba50bc90b80b602a5c5cf4fe68b"} +{"id": "train_041357", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7063e9ffe5a8edd9535a11e45290413fe9fdfc41"} +{"id": "train_041358", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "37ac97a35d136754ca08b53cfb1d9170b20fc688"} +{"id": "train_041359", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "bfb59cef424d18b6ccee9ee36abdf8d62044c1bd"} +{"id": "train_041360", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "46d4eb52fb0755c85957282eb242ed77456259e6"} +{"id": "train_041361", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e902ba1b0f2eb66d8c1704baedecc84956812a91"} +{"id": "train_041362", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "db7b211b8eff1c9fdc6a38473dd6ddcf268df248"} +{"id": "train_041363", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "382d1a9cfb7dee56f61416c3a4f67c6a6a484bee"} +{"id": "train_041364", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0066417c3f0ff676c89e26dc65b725ff224ea9a8"} +{"id": "train_041365", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "d035dee82db109fc575594ed1e7424c6636dbd5b"} +{"id": "train_041366", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "bf6aaae6f5d9c1f6548854ca4b506b7661d4e138"} +{"id": "train_041367", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "6aaa2f460169e406ed31cda342b4e3b82249c53b"} +{"id": "train_041368", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "51d05976f455fa7d085a6d91bc5d1f8ffce14408"} +{"id": "train_041369", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2dc9388e3b9ecab34bc12f0758ab172bef77687b"} +{"id": "train_041370", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "db4a1c71d8007a82495bf5be94a1725ae6686463"} +{"id": "train_041371", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c85eec53c81a657bee363b362bb312988709e949"} +{"id": "train_041372", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "43b3425bb3762346e13d1d6307551be0f8799529"} +{"id": "train_041373", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "5b635b030f4a617a7d52f00b2bdd7650bc84b644"} +{"id": "train_041374", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7d9a5d59f447095d3cc7af8857ec6a944e2b85e6"} +{"id": "train_041375", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3b8735f2a33347a324956873bc68ed71787ba0c5"} +{"id": "train_041376", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a1f0fc610163ad636cd6f6f9b2d595349ff0c252"} +{"id": "train_041377", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "665bf1683979a8bec8ea2cfec724148f973a8e81"} +{"id": "train_041378", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "21d79e6dcfe91961ca8bc99ff7761839afdc4f79"} +{"id": "train_041379", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "18e368b34ec4300a37513ab8fc59ffa53686bca2"} +{"id": "train_041380", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "aa709c61609590a9cc349cd02bbd72f2908cfbfd"} +{"id": "train_041381", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0ebe5138ff9e100c4627555326c11661d3b5a63e"} +{"id": "train_041382", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "8dcb0f5ee57e8746ff2b9c2ea9ff7e0c930c3d32"} +{"id": "train_041383", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "20a82fcd923e7103569724591afbe2459838b332"} +{"id": "train_041384", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "7b7f2fd62ac38334e8113d2f8bf30f6e36fdae2b"} +{"id": "train_041385", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d035dee82db109fc575594ed1e7424c6636dbd5b"} +{"id": "train_041386", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "7461b9ba290695ed3248e03151b7bff56af8cdbc"} +{"id": "train_041387", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e836fc1c13db94105cf1ad1f87e16b8c52681cb6"} +{"id": "train_041388", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "52ec5173f7d7ae6863e3b9902779695643ed7f38"} +{"id": "train_041389", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0650de511e58be140ccf37ef38edb0f049cdf21e"} +{"id": "train_041390", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b0791045ee1043ca69d9be0b45234081e453322a"} +{"id": "train_041391", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d908b9bd6c1d18c929657c909297102b5880d34d"} +{"id": "train_041392", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "055df37c0b1a8a81ca52209a955e4076f7f1d77e"} +{"id": "train_041393", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7e632d121e0fae14cf9deda093ef0340a5226f59"} +{"id": "train_041394", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6b1018b4be3ff30d1af5b70dd896a2e87d021b94"} +{"id": "train_041395", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1c162dd1e875b2a57d20aa4e6eadff3e8dd12e22"} +{"id": "train_041396", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ff3d87ef688ea2e616ff1300bd6e332f55e7faca"} +{"id": "train_041397", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a8329b260324a669c8bb0255fdd00aad0ea4cc59"} +{"id": "train_041398", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fc1bf2551ecd57b96d935fd54d5360e61c0f36cb"} +{"id": "train_041399", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d99352e43216292bde3cbe2c8924dad17c98e1d8"} +{"id": "train_041400", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c64e9d0767e7ab680072a55f896d654477ce0ec6"} +{"id": "train_041401", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "093e8b4a0f49948553a5b303d5d4b759f5ede253"} +{"id": "train_041402", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0d2bad129f144eadb2962dedf741704fdfeeb57e"} +{"id": "train_041403", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "52ec9cfc5507f38985e9d583a402a0cd7394b9a0"} +{"id": "train_041404", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "2d3e9ba1aa3c6c8b3b283c70841cfc0739cfc328"} +{"id": "train_041405", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "bc3b6e1dffc30f3d455cfd928a05769b61f27184"} +{"id": "train_041406", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ba27973c29343c0c15661a9978c91a93a874af0f"} +{"id": "train_041407", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "74051ad0e13869014310bef77457be593e9d26b4"} +{"id": "train_041408", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "16e2614d53ac4707a4ce46a92ebefc665e239f1b"} +{"id": "train_041409", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "43888421ff222fb2058566e0f879d0a908683c41"} +{"id": "train_041410", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8f1bc43f971585de42d3966fecffa1553b4e0f70"} +{"id": "train_041411", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a52aebd151756ba4a1f883a8b5c005cbc5ad86d8"} +{"id": "train_041412", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2970119bf76d4fb7d4a38b4faeb026f50338f440"} +{"id": "train_041413", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "14100f36a452d1249ecb95760e2614388d3df374"} +{"id": "train_041414", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b7ef91aa1ca9591cdea78e42c6703d2971adc8ca"} +{"id": "train_041415", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5f639752b77223f09d2e41780ceb8889ad99bd60"} +{"id": "train_041416", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "438aac7594eb4d0f3c778d76e3f839fff8e46844"} +{"id": "train_041417", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "be19a0ba8659bb34b6c7d0f2a89241eced244f66"} +{"id": "train_041418", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "871678680f794ef6d0e468909380de3d8f272b6e"} +{"id": "train_041419", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c12a77cdecb2d6a0a16faa24ccc8eada62b40c4c"} +{"id": "train_041420", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c067bdeed9d06141c55f87fb4daf861d8290f3d2"} +{"id": "train_041421", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b82fcd95572e91a702f690352624fe19fc73f723"} +{"id": "train_041422", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a7722fd6b2a3bcd970247eaeedd3ed400c92ef9b"} +{"id": "train_041423", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e85fcbfd082bcd639451b226d14654c474be6abf"} +{"id": "train_041424", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "aa93deffeb1cd09f00208c50604c648aab8c76bd"} +{"id": "train_041425", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "0df3a46146a76a151eb645529390292d99a40c26"} +{"id": "train_041426", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "4013265145ba93a56ad6890db0c5016d13e4c3c3"} +{"id": "train_041427", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6751fcda7a1a262a66525e74cb59b82f35a64263"} +{"id": "train_041428", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "91256f7969bb765b60284e8b08710de18f114c7a"} +{"id": "train_041429", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "098529ab3f27db1a061265b23d4e1bd194c5aee8"} +{"id": "train_041430", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0511033023441f519ba6545d78de82e5fd688b20"} +{"id": "train_041431", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "3614b71a194f1862537dc0ccf14fe9125083201f"} +{"id": "train_041432", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6f2be56184e62acf441771e72381fbb58dcdf263"} +{"id": "train_041433", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "45f421a2c0f4e4cb8f149c159c3d6236c2bd19a0"} +{"id": "train_041434", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "42b50fc4714216c3ed6384444d4a13340621e15c"} +{"id": "train_041435", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6baeaa0725de6ec29b70fc4674a5e191c908a60e"} +{"id": "train_041436", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b08b39c4d64d6b987472ad280f4bfd84e0b39caf"} +{"id": "train_041437", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a633f7a7ba9a079b8dc39e812b9764133dae6e9d"} +{"id": "train_041438", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "1aa9a4e30192ad62492d18e93be74c2bc56f51c7"} +{"id": "train_041439", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9acded0aa1d9febc624c55dea674f9fe7f9a698b"} +{"id": "train_041440", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c9d1de993f616623169b9dd14f8ffe9c4f6b8497"} +{"id": "train_041441", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a3cf646c768204ec3c5aded85b4707d1cb585724"} +{"id": "train_041442", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ffa6f9ba3819d99b7a707b743d20967af7c0e773"} +{"id": "train_041443", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "d0f94eccf9bbe0a2814d91c40fe914eb668ff43f"} +{"id": "train_041444", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "bd0fb4bd46748eda2c576839f874e4683a183c4a"} +{"id": "train_041445", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "033f6c0633998d093132fef039289aafa35bd3c5"} +{"id": "train_041446", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f8ca4421cec1b2ce6ab4bb93df21929b7b8f005d"} +{"id": "train_041447", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d1d9e361a8f7700f5d27dae7d3f49eaaa4d10145"} +{"id": "train_041448", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b958b9bd0efd2b33528129db7a1f9cc1640a61d6"} +{"id": "train_041449", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b5f01799fb4ed3fe4ec2432c10003c1314903f39"} +{"id": "train_041450", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8a8ddf954ad2c1c4886cb832e443a062bbefc032"} +{"id": "train_041451", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "dce93b042cfa2518401973623e98b1080781fc8b"} +{"id": "train_041452", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "971e0ca25eb57b7a310cb219165e6b8f7661155a"} +{"id": "train_041453", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b223380cd98aced7f8093deccc6695638110383e"} +{"id": "train_041454", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "25e17c2d65f9697bc2f135265df9c19614fdcab5"} +{"id": "train_041455", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "895d8f33cb6f06508e6f7c1928200dcbe8cece5d"} +{"id": "train_041456", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "2226726b92c9e0677f91c6fd011802b762bf72df"} +{"id": "train_041457", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1f1ca4a63fbffd4f012f3c3d0b41ae2bdfda46f6"} +{"id": "train_041458", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ed5be6ded2541e3ac4d307c961680a9b72050d43"} +{"id": "train_041459", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f7c348b97f6d5532d620541df2bb7f76a0b28dc9"} +{"id": "train_041460", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c55b33bc972f2c7e08aefc5c1e7512e655441759"} +{"id": "train_041461", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e933abd0e77f31c086e18011a9f04389bd41bff6"} +{"id": "train_041462", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "dd2aef1886218eb13503abb88d66eebb237cfcaa"} +{"id": "train_041463", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ad60217dc60acfcdf6ea89366fbf93ec99509a65"} +{"id": "train_041464", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c126946c4b0e0009e975b2834ea1afb1f97c1104"} +{"id": "train_041465", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6daaf2f50e2f2f3620fa5058ac52b32caee5a2b0"} +{"id": "train_041466", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2fc9038f8ef6b2985449766c361a4ba288600c3d"} +{"id": "train_041467", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a95cfb41085dda37e2f109438e9de9dc183fee1e"} +{"id": "train_041468", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "99c572966487ad8f25ba2afe0ddd8660c9272912"} +{"id": "train_041469", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "21f57f818b82078b7b9066b97207d213d98edb27"} +{"id": "train_041470", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9c567b5e7942d76fd0960765a53f9a2b2f749051"} +{"id": "train_041471", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c4023bda6e0b9090701791164e600733e8f434be"} +{"id": "train_041472", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b13ae8cb149a5124d6f71676969e8b73c068ec99"} +{"id": "train_041473", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "269bcf764180e550d2894d87fd208d2e0d4be3ed"} +{"id": "train_041474", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "15e6e179d6be13cbf1dd2c8dba7f0bd0fbc3807f"} +{"id": "train_041475", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "638a76f9c4d2c1b848109f389ab21e3465eb8389"} +{"id": "train_041476", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "995af0f0ed23770c663b976510c55ad11615a15e"} +{"id": "train_041477", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "afbb44e24c93778d7b5ec566707d614d850a100c"} +{"id": "train_041478", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dde118ad6799a5a41839e91d19578c48c41f7a68"} +{"id": "train_041479", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f205f4db6a06c571548a2fe5225dd09dac616910"} +{"id": "train_041480", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d620483ef593c9b2502b358a1f3433e0d12027d9"} +{"id": "train_041481", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "84efb9d2ebaff04b0adbca600392f726b488d161"} +{"id": "train_041482", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "25154fbbb60ab88df634dd9718e6818669fed115"} +{"id": "train_041483", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ac018b34fdce58c50671e2014e9fbeb0cbb4c364"} +{"id": "train_041484", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e443a721c104b977090f47cf7113c19dbbe2159c"} +{"id": "train_041485", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "af3ec344e09530ae956ab75d7298a085447126a7"} +{"id": "train_041486", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "9823ad6ccbdcdcf92f0e2c8f718f61ddaf3555fb"} +{"id": "train_041487", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0ff36caef1c9ea3e8741c284611af0208d3f8717"} +{"id": "train_041488", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ac98339a542c716d97f1670d5b66839446aad92f"} +{"id": "train_041489", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "23d625d82faf9d55335da0b0a2fb8d4a6c4361b9"} +{"id": "train_041490", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bdc3aa61137f0fe4af516526a3f5ce8032515d29"} +{"id": "train_041491", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "4b1bd5b7aaecce47e5d377ea8a762975747e053a"} +{"id": "train_041492", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2e7a64cf2ec31335da1ccd19e0525830adf8180a"} +{"id": "train_041493", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4dd0109fe6d2940e1f03d29af2ebb0ac2b04e999"} +{"id": "train_041494", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "28300f6260faf1b09801a57c171d4e195265a97e"} +{"id": "train_041495", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5cc86bb599de4afe9cf6bd5599fb27766bda4be4"} +{"id": "train_041496", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ada3563da2585f24a95b3ad68b01466d26f12c44"} +{"id": "train_041497", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "50a2151e69cc11854f3447a24af121e54d0678ab"} +{"id": "train_041498", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b22e00424fa1dd4a49ed2e76ca3c883b930415bc"} +{"id": "train_041499", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ed1e491c4172f7fbe9ffae9ad43972ecd0d33ca9"} +{"id": "train_041500", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8d9d2d24abdf50f827584a0f2134d5ee6c63ce9f"} +{"id": "train_041501", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ffa5e5eb3c22df65fbc7246e8ff0127530e5b3e0"} +{"id": "train_041502", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f6a5221d9ce41bb00653d5d99d37f4e6194e32e2"} +{"id": "train_041503", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6665d28c39ca319aea392b24f7bd54d149ee2f4a"} +{"id": "train_041504", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "73a21362c45f97cb2f286b729edddad0cf4c3558"} +{"id": "train_041505", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f700eddc174448d3d61bc8734d5a38d409f9c6cd"} +{"id": "train_041506", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "7f6acc6c0acb1ba4ee8798d5d39bd9c9967117a4"} +{"id": "train_041507", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "249770db57435a2b3b157a9e7fcc8f6c7e517e78"} +{"id": "train_041508", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "4e561878b179666fea17bbf89aa4704abad9d114"} +{"id": "train_041509", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "3c9751395eb265701c4ddf0eade594f6854ab6f5"} +{"id": "train_041510", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "348c0260e13091705cabff1c6b69311be2ea868a"} +{"id": "train_041511", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b8d93fc726fa5a5b3f5ee792fa718b05681e234a"} +{"id": "train_041512", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e835241685c4476491613698d819e3db804edce8"} +{"id": "train_041513", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0593bc73c186c92188defdb6e5ace4afb62523f2"} +{"id": "train_041514", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "518ed5f92eb9c52ed1ea5708082e9b2ee9a14d1f"} +{"id": "train_041515", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3712918e80ed5f05e1aa3517d285cd269e38cf5f"} +{"id": "train_041516", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ce189ce4dc588950ae278e1c0792f16ada8edbb4"} +{"id": "train_041517", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "47278affa44a869f9af494ba7ee2a15241d01840"} +{"id": "train_041518", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7c1ea7650e87265110f7a629f0fc5c9c04c5d7b5"} +{"id": "train_041519", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0d03cb07b37567fef4b6cae5ef1f36be18dba261"} +{"id": "train_041520", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "792b6fe90686b7d0ed4fba7527ec8d2d2541998f"} +{"id": "train_041521", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "11f3c74e0c57a8bb075906c7c6ad33efc4feafc9"} +{"id": "train_041522", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9f1dd16e8a00bc77444d33ba406cf8d6fbfa61c0"} +{"id": "train_041523", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f64dc1892e34f9171c2163849f677c50445c6159"} +{"id": "train_041524", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8b5988304797ffc966e7e66e494f6fefbc2b7d0a"} +{"id": "train_041525", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "12fad59f88090764d243553ae4638828b296fede"} +{"id": "train_041526", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0b59ec71592723259a9c69728ba5b2790617fc59"} +{"id": "train_041527", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d021c4553b9fb09a3b7e2b1fab24d578e5650212"} +{"id": "train_041528", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d78c8b01cb4a895730a0b3184b7f3340f3f69ae8"} +{"id": "train_041529", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4e75d8a1027f37af352fa56204cbaeeb81a98393"} +{"id": "train_041530", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "cf8a1a8e1ee17f230cf53fcaae829480dd656641"} +{"id": "train_041531", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "acb8e0e7c039a0e8e52d33102983847a24272391"} +{"id": "train_041532", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "00002db1ac8f904315b39eba6376ba8cb73bba2e"} +{"id": "train_041533", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "fa78ff8da5f99d037ed78cfcfdb10b01ea6eb783"} +{"id": "train_041534", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "861b73950b0fc596e25e0e6d34fef07e4a7894a1"} +{"id": "train_041535", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d0d16036196fa7a9c135451358c2de75c6bf3ffb"} +{"id": "train_041536", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "75d1b3a34f8f5e9d0623473ecea1df0376f25dd0"} +{"id": "train_041537", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d4d4b2070aa2c81c2c53bf0620d63bb209cfc5a4"} +{"id": "train_041538", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ed8860354cbe606a8c6c7f9a06cc1805b4f38de9"} +{"id": "train_041539", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "df549922e062460e57d49e5e316be49a24609f15"} +{"id": "train_041540", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f306a29ae3ec4200397e5f038cc1d660ca852d1b"} +{"id": "train_041541", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4675d5592d045fb5f671f947ff5184b8467455c0"} +{"id": "train_041542", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8f0c1da27bbb58b85c480276a97682074363373e"} +{"id": "train_041543", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "77e2e402c1d3dfaa5a824daa4cbd15ebadad75ce"} +{"id": "train_041544", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "15a499cf24c00c94b35f0c7367e0636d1d9be2c0"} +{"id": "train_041545", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c448433a19aa62fbb91a7a3c5e2828f5ce73aa20"} +{"id": "train_041546", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ab1fdaedddad235fa49f1d0d0e016adfc3be3c40"} +{"id": "train_041547", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e588ef1ff8be9adeb6b3327ed76fdf2ad26a1977"} +{"id": "train_041548", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "0524c95c6cdfbdd4c48792072b792d499b2b7956"} +{"id": "train_041549", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ffc09fa4e59d8d3fd8cfa86fe0bf300189936097"} +{"id": "train_041550", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6f011b2e04fbf6281040b09b4cd3af5614bf77c3"} +{"id": "train_041551", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1dfd0ec7977113c3fa45c1052fea170dfbe71ad6"} +{"id": "train_041552", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b8bff98391a55a208a92582228352d4730698076"} +{"id": "train_041553", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "539efebcf19c424eeb67a9fc614083ab2f30c146"} +{"id": "train_041554", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "93b341d3ea3406b577868f7f474299562e9dbfa6"} +{"id": "train_041555", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a58447e0c4af391e22bf8f19e23da6ab91ea1e10"} +{"id": "train_041556", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "84903fdf9642cff3bff2017230dcb5e7409714c4"} +{"id": "train_041557", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "38a0a897c1b5cca0297c87428d2df1cef5949a13"} +{"id": "train_041558", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "22e5322b20809d42fb090c25811b1416d0949d9e"} +{"id": "train_041559", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c590b797140c7f138e407fc96349c866dd4e2449"} +{"id": "train_041560", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "bb36504a98965758b3b40784ea823a6fdddf8d57"} +{"id": "train_041561", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "79e52883972f69aa1d3df7ee4ae46a4e13c527a7"} +{"id": "train_041562", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ecb28575a629127bda3cccf8c8ae8fd54737156d"} +{"id": "train_041563", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e8784310f9f76fafc0125b6e294088e593330891"} +{"id": "train_041564", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a77eec75e3e385301796dc2a75e143b84edc64c5"} +{"id": "train_041565", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9c4f090499e344ef3e40cc4e57ec8eff221e2165"} +{"id": "train_041566", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "dbab826cdedbd41378581fd1ad53ca0295a96580"} +{"id": "train_041567", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5bcc51b2554c09e4eb4ed70e5f4de6f5fd97533e"} +{"id": "train_041568", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "87ce239165d9a1a315f243ce89ac9509652a2d2a"} +{"id": "train_041569", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6a356a05620803994faf49878265f2c27891fe72"} +{"id": "train_041570", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "362f2f9f36070cbad93b6f436e090a456c049660"} +{"id": "train_041571", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4a6e843446c41ca468f794d6ef30cf448d60cbca"} +{"id": "train_041572", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7ca5675168ddf625bbce98f66484dcf2cf9c9821"} +{"id": "train_041573", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2343eb74a67e533963397253dd1e4bfd1e5fc320"} +{"id": "train_041574", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c42b10818464b0bb07158f14e167b002b5e29c64"} +{"id": "train_041575", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f66d154183ab58c869afdcf3a66e7edf93ce0f23"} +{"id": "train_041576", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4982a5c546747167171e6cac20a0485a6acfad00"} +{"id": "train_041577", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9b4514bb2102056aa9631adb732eca1a1d80d458"} +{"id": "train_041578", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "43436cff6b6be1dfa9183f9debdcd1e96b83b121"} +{"id": "train_041579", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7ee0ccf6590d06acbe50591dc8d81824c17ea14d"} +{"id": "train_041580", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ec9881ce399e1e213c5f6bd091a308fe291e0291"} +{"id": "train_041581", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4bdfef78a48acbed494bcb74ded716bdf7bdd9cb"} +{"id": "train_041582", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9b681f0a2f1a2fe37616c5f1db54f41db378a12d"} +{"id": "train_041583", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "483319a9808d023927b09d94bb9ede9f78961492"} +{"id": "train_041584", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "554b773861acbc64e22a039b70cf8ceae6c72f28"} +{"id": "train_041585", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6e08e692bda81c92588e587e67e4b6c3e3045f3f"} +{"id": "train_041586", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "810dc497ed0809eaecdc07de3ac749c1a4863c21"} +{"id": "train_041587", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "504bc69cc928ee2a32d52d22b4f362d5ca219b24"} +{"id": "train_041588", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "78ee7058c1e51ec000a1388aefe46c00eb0aae9a"} +{"id": "train_041589", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a8e329463753380105d807d6bf02ce2157f11592"} +{"id": "train_041590", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1db480612a4866b831a912796425cc5d75036f80"} +{"id": "train_041591", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "49c73d1159195c419099e02352fcb20bfcabec12"} +{"id": "train_041592", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "62e86d2da29f02c81bace762f26ae8e5bc73812d"} +{"id": "train_041593", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dc3a2d58f7fb0baae7063f1f00cef9ebb4d84a40"} +{"id": "train_041594", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c24bc02c984a49b407300dd57f56998e74bba85d"} +{"id": "train_041595", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "52db5b2ebc68a39c76a072e6b7c91ed63b6e1400"} +{"id": "train_041596", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e14f70e8cbed6e9d38767a020929511d1c8e4566"} +{"id": "train_041597", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8e5820bd57d0c79c88b21e454961a9afb1d8de24"} +{"id": "train_041598", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f2c20d2c67fcf4c14db4ef31acf1267c203caea3"} +{"id": "train_041599", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e75a9755a29ef0b0c1f81fdb8bbcb99cf7edb40a"} +{"id": "train_041600", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "02a94d2513ecd7bc9da8342e0cf614cd5dc79cf9"} +{"id": "train_041601", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c6953f7c6e2d7532e2458cfb457913ebf6334500"} +{"id": "train_041602", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6513217c59cf375bbf9085abae11ea313cfd42b2"} +{"id": "train_041603", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "fc25ac099ca089f646109e2a094a067dd3b89e84"} +{"id": "train_041604", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4e6fa328f1a417265c8ea73a5740e1af22e6c897"} +{"id": "train_041605", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b7708d2695534a5f05a8fed59fc67d5c23153a64"} +{"id": "train_041606", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "27bf7efb4f7e47435294fc3726a69aacf8de654c"} +{"id": "train_041607", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a599e6dc0f173b20d0c9ab14aee723faa3d33e5f"} +{"id": "train_041608", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5592fe746dcaca494bcbe15096151f13e652cc2d"} +{"id": "train_041609", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "3bc361eaea596d74166bd450ff90e317f2b72dd1"} +{"id": "train_041610", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c676a9bc9294eecea9e61f08014868c685e12ae4"} +{"id": "train_041611", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5bc762f153589aff60f5b8105c72ad7d98256507"} +{"id": "train_041612", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "08582f0f29c47039a8560ebee40663462ecc84e6"} +{"id": "train_041613", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0a8a7f385fedfbd233a59e61ee590b82e3389639"} +{"id": "train_041614", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "016ab9ffafecbab94afa1f2f12049015e179b480"} +{"id": "train_041615", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d87a7ca998acc166ef4df3ddb672cc38dedb316e"} +{"id": "train_041616", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b4119fc3b1b0f048a7cf9d960266f5d243e2971c"} +{"id": "train_041617", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "23d625d82faf9d55335da0b0a2fb8d4a6c4361b9"} +{"id": "train_041618", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b8abcefa71b25ce959b45e7a669a6ae46c8b033e"} +{"id": "train_041619", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "b2bcdd4fc0ab51b5b3908dec874e39497651fc09"} +{"id": "train_041620", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "133b8c60eff95201c977079c5dc6e4463324a8f3"} +{"id": "train_041621", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1cde996d8a4ac5f366b6978db318cecc440e0a2b"} +{"id": "train_041622", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "42dd6a8d5d9c2c50380e1cf555c9d2e2c72139c5"} +{"id": "train_041623", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "4799ea54076ea39dcd095fa1765c5f2c12c55737"} +{"id": "train_041624", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "13a6b5c6861160c4b5860b35a0d0c8d49f8cbebc"} +{"id": "train_041625", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8e8c15786b04d4b864e506efa99b8ea3379c818f"} +{"id": "train_041626", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "268bbcff3fc7e7d43417528030467529d89a5832"} +{"id": "train_041627", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "db998cbf63f233158924add6e9fc00aad233a801"} +{"id": "train_041628", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "416d88e4289a63131e64810458474864a6d275e5"} +{"id": "train_041629", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "88bb11f808244efa9d96e6d8dc49a1d7257d12ec"} +{"id": "train_041630", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "477bee63ea1a8a47d62753b6337bb25907ccc083"} +{"id": "train_041631", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f899c28e1391de17c0a47b8690b128bb4e0352c1"} +{"id": "train_041632", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f0babb971f52704f9392cb4455becf808eddda81"} +{"id": "train_041633", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "5930b51a4c910164becdc40eb9b639c98ca386c1"} +{"id": "train_041634", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ed8c88f406e9855d9bff81fafe19b61f7cdf8bce"} +{"id": "train_041635", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "daa456c846956f86dabfa9fd2a6ead94186aa051"} +{"id": "train_041636", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "27bd0ee398566ce6a624d13a7127aa230c727a8c"} +{"id": "train_041637", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ae8a2b51c331a10627ed35cbe3800f999eb95b14"} +{"id": "train_041638", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b600fe45b894bb4b6f2351e78c52576dcb0aba32"} +{"id": "train_041639", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "838259b8137ad6b2ad616f2e529f2f52f9b836b3"} +{"id": "train_041640", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8a10d9487e962fc5d16681942297f809123f495a"} +{"id": "train_041641", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "280ca86d910435cd53b669318e39c3d34870e842"} +{"id": "train_041642", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f95d63ff6beff89293bea6990f2b1fd12167dac7"} +{"id": "train_041643", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "11d52b4734631b0fde38231f4c06156a0f19df01"} +{"id": "train_041644", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "845871192f56b0aaca202160c70c9b41c2b4ff3e"} +{"id": "train_041645", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "445994dfb7a62391e6095cc0928b023e61f20804"} +{"id": "train_041646", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6b077c53eb58ca22de3382578910911d195a3263"} +{"id": "train_041647", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "596df33c9e73096610df1efe8d8a34bba1de3179"} +{"id": "train_041648", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4a7aebb47b959412f19429b5c302c688bd6069ec"} +{"id": "train_041649", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d721870d8b256604a55aea6250f4d82af59d9a72"} +{"id": "train_041650", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "117f413bce7725bddda8ab90a5fcdeef615c3122"} +{"id": "train_041651", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cebe87c5f1d9c5d765f6eab3e8a711c5ed12e1fd"} +{"id": "train_041652", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0d6ebbb5efa3519d12552de0d9d68be7fb93a587"} +{"id": "train_041653", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3b214d8c3d0fab7a440ee93fbf8291853135e7d0"} +{"id": "train_041654", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c2996ff23231a0abcb30598f0fd52d2a9251f731"} +{"id": "train_041655", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b44f81323d2cdc66e23b1e6c449d416457842f04"} +{"id": "train_041656", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "1cbae78cb4ccd7077d97b74ee7bbfdc491f7babe"} +{"id": "train_041657", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "16c4879ddd4deb99720f0f6ba5e1e1df4ae14d1a"} +{"id": "train_041658", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "cad30b2cca704957706e2035f2734f3ce714a61c"} +{"id": "train_041659", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "8785c4edb3e994abcf85dfda967f4df8ae83368b"} +{"id": "train_041660", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5b1aa6203dbc635c0330545719cf42c849e29eb9"} +{"id": "train_041661", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "08e11a035eb6ae16f2d82da1bfee3571fafa184b"} +{"id": "train_041662", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2d8d97c57eb21b71a06a9b35087dcc9e6f6db6a6"} +{"id": "train_041663", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7b8df5e6caab5e1a62fd2c901abf3b336134bd9a"} +{"id": "train_041664", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8e279c74da6888aed7674018d55fe1fd212d64f0"} +{"id": "train_041665", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "40edd0e53910dcfec4e99477d7c0762c169f9bac"} +{"id": "train_041666", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "bc56a9afcef3ea7254ef6a760208f2ebdd51ad21"} +{"id": "train_041667", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f656a614038a016cf3f33b8b38af0b93389a1d8a"} +{"id": "train_041668", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2e376b493a22c1486a74ab5ab4cd1ed752642aa2"} +{"id": "train_041669", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2603b228f88c9d8d2717c305f47b132f1ef3fb92"} +{"id": "train_041670", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c935343d1c7cd459693183fae8184a20dccead33"} +{"id": "train_041671", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ffb4ed3f0bfdef16fec32ec3ddc3fa3d50ae441d"} +{"id": "train_041672", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "fdbe93168f20013feab53284a8285f20b0f079d6"} +{"id": "train_041673", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "74b7766143b188da1b2053f36f37f514681c3daf"} +{"id": "train_041674", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "91d503a7b28a21c3b2509d50345f5ce8b6bd0611"} +{"id": "train_041675", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f734ed13ab0c101608ec2e020cbf34a7a916d250"} +{"id": "train_041676", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "44ba2e178e225edf648639a6623e94acbb1a03ba"} +{"id": "train_041677", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8e279c74da6888aed7674018d55fe1fd212d64f0"} +{"id": "train_041678", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "30d8f1428c15c8a43406b2dc4c726ba33972fa96"} +{"id": "train_041679", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0c933b349417011c3d5824e5cf9633c06c352ebc"} +{"id": "train_041680", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b0af7c43b056d3ab8aa36d82a8602a3811095376"} +{"id": "train_041681", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "98a4e9bfcae1c230fa455f3ab5b21ea2e7469fd3"} +{"id": "train_041682", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "577e4c33ba94afdba82975f04a85a11612db651a"} +{"id": "train_041683", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "d8be562ad1d147825a64603317a2d071ee7bddf7"} +{"id": "train_041684", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "472cea45bbaa6859417ce9c5662fe6724f979359"} +{"id": "train_041685", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "04ae405434bc1efd08adf401f29154943210457a"} +{"id": "train_041686", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "41b02d05f06563037e9e4ff081e9e124c67b26ce"} +{"id": "train_041687", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "944ce600ddf06a66c1ab7947e9642363d359bcfa"} +{"id": "train_041688", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "5a694566098a938644b6abb967d25fda71463248"} +{"id": "train_041689", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "31506cb1d0f68317059ce232557d0a6718a8ed7e"} +{"id": "train_041690", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a8627fdcb8419f4cb591220511f977ca3c71b72c"} +{"id": "train_041691", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "678ae6324e32b2ea92e1005605a87384f785fa4e"} +{"id": "train_041692", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e565d5793a4a21c459bfed9d353b035e7a3201e3"} +{"id": "train_041693", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "afdfda85759e1f9b9c0de34be96407e33c11b688"} +{"id": "train_041694", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f078cfcec33b3624d105cbd880d8f2f686198973"} +{"id": "train_041695", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e0119bf55b836e6cc9296333694e50c638a059a7"} +{"id": "train_041696", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c7121a68b006df21c00137bbecaa571e32201d1b"} +{"id": "train_041697", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "559a01edd34635576b5305368d0d64beba01ee62"} +{"id": "train_041698", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8bbd228f493430a2cfdb672cbb19d3e6e66a0e75"} +{"id": "train_041699", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0c1f06af552e67875c73e60bfd4d7fdf6ca0be4a"} +{"id": "train_041700", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ac8b21f0ab088629dd4c897a0d58fc014465dbe3"} +{"id": "train_041701", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "af0536687d659b46bfe66aafeaa9644d974de42a"} +{"id": "train_041702", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2a4c28f357742b6b873a554a993399be62ff68eb"} +{"id": "train_041703", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "66c703ddfc0083e0cb53d39e6276e98868d6a17c"} +{"id": "train_041704", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2929cabe32f062941be24a63110b5b147fca0969"} +{"id": "train_041705", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "48b76ab28adb8c067df9be66090530c79651938b"} +{"id": "train_041706", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "45f9b0c296e1bf4ca0972d9f7d6f894a1c182071"} +{"id": "train_041707", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "9b730f619bf08193001d22c33395505617146772"} +{"id": "train_041708", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "15dc094406b7805b28b92721d2a765a00337fd46"} +{"id": "train_041709", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "184d5f486d243a6d163851c353d08e963a9a25bd"} +{"id": "train_041710", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "58ce7e391891147164f8afb45c385ef787d71a92"} +{"id": "train_041711", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "643dd91f09f8b50c9b9915f411cf3eb2928fde29"} +{"id": "train_041712", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7f1b0a0340df55c00d4ed8e9fbe79ee10f395ef8"} +{"id": "train_041713", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "99485c937a5629aff90d6ced9fa9848c76ddcb47"} +{"id": "train_041714", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d1a2922e625b788bb94c56d26466fd2bcf2336eb"} +{"id": "train_041715", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "06365a9adaf46dafa457b286eadc9d2e5a12cf57"} +{"id": "train_041716", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "877dc78d2a073d07b52bde906c362568293159fe"} +{"id": "train_041717", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e7e2b96257fe453e8f1b78dc19e1b60fa6abff45"} +{"id": "train_041718", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "175a9e4f319e89d6c8e3643d89d52692949bf0c1"} +{"id": "train_041719", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "801f015235026752b493e3c1960ae6cdefa40537"} +{"id": "train_041720", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "658972004aff3f8b86c6068822ea15c82bf2cf18"} +{"id": "train_041721", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "aad9d5b5ec22c346aef8536996d2eaa4e8b823ac"} +{"id": "train_041722", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "203da55c15dc1805f216fed287a6845d8e085bb6"} +{"id": "train_041723", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "c81c6b9690addc753035b1e9f1679b2ff6f02f51"} +{"id": "train_041724", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8a05cb3516c0edf553ab7a10d4f962c6228c053f"} +{"id": "train_041725", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "baeba7d5bfbe388393beced0ffaa00c3b3cd61e6"} +{"id": "train_041726", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "063468880cd23cf11bbe41cfdb3d836311d13be3"} +{"id": "train_041727", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "244d7cb5b2fda6de457112bcb252ebb7e3991155"} +{"id": "train_041728", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "cfb18bd5b0d09644057ca0eb40d569540e648422"} +{"id": "train_041729", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3743d4d74c2633542bb6727831d6f49c8ee47e0f"} +{"id": "train_041730", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "15dcc04785feb6229b3f773c523574f0498aee6e"} +{"id": "train_041731", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "accb22b201bb1bc0dc27232f076015ea156b1684"} +{"id": "train_041732", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6540380f6c2029a952a3d58d80616a3361c02f03"} +{"id": "train_041733", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "abb0ec6e65f5985879b6208d86d14c1f7834b2b2"} +{"id": "train_041734", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "19f6abf555f095f87c55e17b0ad10dc890dd91c4"} +{"id": "train_041735", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c38758488d2e22a60a6ba77b10ce122a5017b4fb"} +{"id": "train_041736", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "76c9892c2bd8e61f0838cae655deb0d9803afe1f"} +{"id": "train_041737", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "070e188c6570158f8d63dbe847d312397e9fa24e"} +{"id": "train_041738", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2ce4acd63e113825f2f80f4646f57e3b72669e5a"} +{"id": "train_041739", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "42e2570230da7d06d0b5713fefa827d15f60d3fe"} +{"id": "train_041740", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "10afb4c993b60ccf7562554e609b669d7bb93d2d"} +{"id": "train_041741", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "346f2887269d782207b15ebd39fd4f6e35cd35f2"} +{"id": "train_041742", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "41fa997f95f073d97eff1b833f0404fdf5d4834f"} +{"id": "train_041743", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "601fad7380f323cfc2beffac7cc2a5d66358e00e"} +{"id": "train_041744", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f97ea717caa30943a7bded196b55b56f9336c82c"} +{"id": "train_041745", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "bf8eb84fee14432baae52b3373ada4ccd58e8501"} +{"id": "train_041746", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "412cb12f44b0160ee3a237766f23304de63bc3da"} +{"id": "train_041747", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "cf9d8a803b5055967a1c16ac4ffd69367a3cd6b7"} +{"id": "train_041748", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "21d2e64c08f7781d3b35db7d5de83a1f3c26060d"} +{"id": "train_041749", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "923febdca577b74f2e0a0b2a13e98da1b2c85787"} +{"id": "train_041750", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "407f3f06711a590aab14a933c288ebcbfef41a12"} +{"id": "train_041751", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d8a2b82c1a4e1754d8bef63dbfba127822cb168d"} +{"id": "train_041752", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5160a31db57211ea02af3e7c1f0249a1b2c3db1c"} +{"id": "train_041753", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "9203009c6691a89cc380f60cdfdbb4d8b5ad2577"} +{"id": "train_041754", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "9326e1c8e4c8784a75badf77ecccf98c62582f97"} +{"id": "train_041755", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a6b9a7d9c3d0988a58c2dfbbb757a99053f45e50"} +{"id": "train_041756", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "edbb5e1213e058fc4f5d5f14159ceddf5a5a86bb"} +{"id": "train_041757", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1794833462fc090696c87076a3f8267fa447178f"} +{"id": "train_041758", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a7c6faa909defefa3f77bf934f3ba054f243e215"} +{"id": "train_041759", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "4a63b7b6b225dcc0edb03e8d9906a683e72dd24a"} +{"id": "train_041760", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "346b82f194bd17ccc37d899f0903e1cd283362f0"} +{"id": "train_041761", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "9ec0556e6dc16793a8b47e7bdea32dfe71810d38"} +{"id": "train_041762", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "17df15fc5baf68e2b908e897c9d21cf6e6e8f916"} +{"id": "train_041763", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "da6bf57eee700dd5e4a137568241d6ffc02462b4"} +{"id": "train_041764", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "57adf6a11f29a3aa620daa7ff32d4622d3ee0313"} +{"id": "train_041765", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4adfcbe001b94dd4ba7074134cc829677124ccb5"} +{"id": "train_041766", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "29473671db3a7b286f0b0bcf9a0e8f490f87adb1"} +{"id": "train_041767", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "087dbe3a56c8c5dc7e4c448c527ac2d86299a1ed"} +{"id": "train_041768", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "95444ea83093dc8611f6b0f725ea557aadada3d1"} +{"id": "train_041769", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "318bfd5c4e97540d974f1cfd2829434b1e401a54"} +{"id": "train_041770", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f269ad566856cf4447a2ffadfdc258677d70170d"} +{"id": "train_041771", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e713336670c918157071488fa943cf466db375f5"} +{"id": "train_041772", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5a66741842b1bfab1c378bfad8bb68969d63f332"} +{"id": "train_041773", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f7617a395a020caf01bf1a8e5e5a0fd344c62731"} +{"id": "train_041774", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "cb280245a740cccb8fb6a972ebd29b59bdd55faa"} +{"id": "train_041775", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6615c83843b2259fb2c9be3b97b91deeb3991cc8"} +{"id": "train_041776", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "73c58e775427747ae08650d459378182dad7ab0d"} +{"id": "train_041777", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f1e864c1ceefdcca043d93edc7d62de012c73797"} +{"id": "train_041778", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7f2f6da07fcc6103b7535aebb9d2c8a4541681fd"} +{"id": "train_041779", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "baf86579c811b369754183c5c795d3669fc5d89f"} +{"id": "train_041780", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f29fe2eeafde04a37f274cd193bc15392aec420f"} +{"id": "train_041781", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "809cdf85d6fe167ebdb99e1bd9b0fccff576a5a5"} +{"id": "train_041782", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "bac31559ebd094d9799770a21eeaaa0dd93e7245"} +{"id": "train_041783", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e004092e08018524b2cc09546937d63ba454e59a"} +{"id": "train_041784", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8f0c1da27bbb58b85c480276a97682074363373e"} +{"id": "train_041785", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7c8f21470d5740754f01f25f82b09517ebf565f5"} +{"id": "train_041786", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b49c8196ad50c1f4bdd87779bcd60881f8917496"} +{"id": "train_041787", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f4eeb7eba0d828a1f453d73e73d96e5cdcc0ac34"} +{"id": "train_041788", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3a073500ff6b8a1efa524f51c57c51b743ac7544"} +{"id": "train_041789", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "b489e50a85a378fbe0b9a47cec0d75173ffc8f7e"} +{"id": "train_041790", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "55d6393bc9b3fdb453f87cf9effe84526a989932"} +{"id": "train_041791", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b8468d58a956a5877081f619e128725b2fd35ecd"} +{"id": "train_041792", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "be479522072426273463829720e9f33b6e126230"} +{"id": "train_041793", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "14f4f6b84fdeb73222ad7bae75a464d78135040f"} +{"id": "train_041794", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "55ac8e868de881c5f4e444008b58ca2a2c053aa8"} +{"id": "train_041795", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9541e28e965f86a1386a3001acdcf2b3697f9050"} +{"id": "train_041796", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f42f0c350677cbedba664d6d8cb31d65b43f4eb9"} +{"id": "train_041797", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "de22ddf12fbbbc0ea1f66f0b5d2038ecf3d53583"} +{"id": "train_041798", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "cd6c80b8bc9881cb7060b2da9c1d44c83dfd3190"} +{"id": "train_041799", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "29fdfaeed9d10c2d2af70be9bf71333ae926c2f7"} +{"id": "train_041800", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "6e9d2d15859d8e3fb08895e6827842c3f60dabb4"} +{"id": "train_041801", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "e2945353561f1337c3080bb7ee12b6d4620216d7"} +{"id": "train_041802", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "51bf39ffb751a2e3a2550e94b6a00d6be38f3b17"} +{"id": "train_041803", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "73ae9b4fd08cb990c3d96975faa2c026997b6244"} +{"id": "train_041804", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "95cac0748ce7f405582dd0280a5dda37f98eebbf"} +{"id": "train_041805", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ada76ed9a7c2d130b1c52f9e0dbdc12550dd7721"} +{"id": "train_041806", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c95703da04960ab464f1bc2078d2bb27b721e88a"} +{"id": "train_041807", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "87eef2bcadfbb717b2166fcaa1c6121982d0c51d"} +{"id": "train_041808", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d992fbc9c45d6e168e8b57b53d6c3d57f13bdfb3"} +{"id": "train_041809", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ac6453bd6eac93869cdcf7b4d09cbe8082c168c1"} +{"id": "train_041810", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "84738828a9af27008f468654ca76ad6e56266454"} +{"id": "train_041811", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3a12fbd4449182acbfad259166c48ccfbdd19942"} +{"id": "train_041812", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "805f3629596a6871dd5685eed7c40408e3688428"} +{"id": "train_041813", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2a9568efa981f0ab927e0f2d766882e897f381fe"} +{"id": "train_041814", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "de60a364d6e6dc7def2621e8d1729b3d846fe430"} +{"id": "train_041815", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "98a61d6471e313f946ff4b32c250cc26cec0f0c1"} +{"id": "train_041816", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2ef3748285da8790cd5709f6c6e0ccb4745f1349"} +{"id": "train_041817", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "28c219244247278d4aa622280d4a3e77588ca4e9"} +{"id": "train_041818", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d28074892ab4fb82a3bbedcf99ea2ed9b148abdd"} +{"id": "train_041819", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5de02970c29173bc7631549de69a3c6b19594f90"} +{"id": "train_041820", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5451324b514c92101126595dafac1c7ba50ef8fe"} +{"id": "train_041821", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9aa31ba152b4fce020b1273b4eb16671d4b0a6ff"} +{"id": "train_041822", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "bc1ebf0fbed6532bda1f06c679fb0a72a9a367b5"} +{"id": "train_041823", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1430fb6b7ce7e2cb070f66ba05329cd6febdca7d"} +{"id": "train_041824", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "51c6b0a3f36ee5ffd441fd9123f51a0a33da46f0"} +{"id": "train_041825", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "17f6f2235e8ca97ecd55aac6321075fef4117954"} +{"id": "train_041826", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4f3d6c8f60a27abd7272658f49851fd997ff49f0"} +{"id": "train_041827", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "abc5e3b413f45018d8e29dc476c2112727b2382a"} +{"id": "train_041828", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "6f2d7ad61bf7df85c51a883b950b80c357bcbccf"} +{"id": "train_041829", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0f6f260a37e67301165692a565735807eec1af8e"} +{"id": "train_041830", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6b7feaa9a3125626eb2a5e80d4e76c1dfa8a17f6"} +{"id": "train_041831", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "7b622aac026be1f5b8bb4081ac71d78087b8245a"} +{"id": "train_041832", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a07eb14b4a9fafb2b064b5f263c56a7132c7539b"} +{"id": "train_041833", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "4068a1dd4a6e7f294afa29f2c1be297f8832bd27"} +{"id": "train_041834", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ce09cb16c44bd1ae8058a8e5c4132c9efb1e2bc5"} +{"id": "train_041835", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0945cc6764131a92df86f654874d1b9829b76a1a"} +{"id": "train_041836", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7bcedf4228cd0a5b721b60d8da8721976c831d16"} +{"id": "train_041837", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "9a6f7b2aa52e45f656eb3185b57f0350c9056afc"} +{"id": "train_041838", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f77b09e98021c51d2c1a6430564a1a7be249cc08"} +{"id": "train_041839", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "bdc384b230090c50841f7b05aae667bb85e138f1"} +{"id": "train_041840", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7e93bc51aafbd97ff064305bc65870558919b5a0"} +{"id": "train_041841", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "7156f86ebf1d0606420f355b22c6469da0cbd612"} +{"id": "train_041842", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9a3fa15cad0888d8b3a180ba31bfb2a3dc22d435"} +{"id": "train_041843", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "df856c34b8acd446b0188b4847f43fa0830505d1"} +{"id": "train_041844", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b2a01d1cf5c9c0d4776d50464098e9919ee9bd58"} +{"id": "train_041845", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "bc56891e8cb15b074e9413e70c8c6a99c5cac848"} +{"id": "train_041846", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "df6b45fc4b3020d3638168294ccc0038a0769669"} +{"id": "train_041847", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1b4da5c481ac1fe414b0671561937a0ffdb3955d"} +{"id": "train_041848", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "cc8a4a0f4e28241da0c5a7742577d9bfdbe9d590"} +{"id": "train_041849", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f299e73ba46426c5ba7c19ad65b0cca050d2fdc8"} +{"id": "train_041850", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d34566264bd00bc4f99d9b559fbd311c33256892"} +{"id": "train_041851", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a5bd3971c3df6c94a3f8653c303257d66253fdcd"} +{"id": "train_041852", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "7ee0d26a8c38b42030ca5a8b56a05f57752a69ba"} +{"id": "train_041853", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f1ae743d374f8df4f2efb517701d8adec67248ed"} +{"id": "train_041854", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2059e5c1317333502c960a79117e7ca8a46308aa"} +{"id": "train_041855", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ad8cc31899e0fc37de400aef675ed4420fcf681f"} +{"id": "train_041856", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6feacc8e4c0b3af3becbe550f9afb1342c64edf2"} +{"id": "train_041857", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "36fc18555d6145d91522acbf18935d5bb6db4db5"} +{"id": "train_041858", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "017535ed54e36e4735cb6796f254fa0bc2359912"} +{"id": "train_041859", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2b3c52a422e233acd9ef2f41e7271b728c87aff8"} +{"id": "train_041860", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e9e7cf7a9cbe89fc533539f30b7624bd2114ccdc"} +{"id": "train_041861", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "1fbf0f51b8e92477e0af86590451363569914eb6"} +{"id": "train_041862", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3a516dc5064803966a2d034f3c87f62578e98596"} +{"id": "train_041863", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "240462aaf8fe71b56565a10386e39eea42511e19"} +{"id": "train_041864", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ee153d15a087da06f55037969e5d09e64a25d6c7"} +{"id": "train_041865", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0a9dcd82adf6f9897f5357e6abd70a18a9aae2f0"} +{"id": "train_041866", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "85679a5a1295e460457f6fed31e7a838287b1581"} +{"id": "train_041867", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7b25e8f5a65c7fa5d457e8a6809dddd9d96a6355"} +{"id": "train_041868", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "72ccf14c51952945d265ab469d318aa926b44b80"} +{"id": "train_041869", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "cf8a1a8e1ee17f230cf53fcaae829480dd656641"} +{"id": "train_041870", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8610b88a510e4d433af73a2f38d49160efb6c4b3"} +{"id": "train_041871", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "095c9de977249ecea23b0e24729cc9aa331c8746"} +{"id": "train_041872", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b391c651ec9d5e9580d454005a6cc3fad2731a7a"} +{"id": "train_041873", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "69ae89eac6107085e936ccbbe90fdb6945169847"} +{"id": "train_041874", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "fa60158e5fd2dd17741b311d2f08db858bf43fed"} +{"id": "train_041875", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "13bbb3c429c111f2f250c6a676faf2b33ba58932"} +{"id": "train_041876", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2415a58689071dd3f70fda78f6e11d0dc5f06475"} +{"id": "train_041877", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9203009c6691a89cc380f60cdfdbb4d8b5ad2577"} +{"id": "train_041878", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c32bfa5002cb0f3f5c2b64f67099e93e23ff1274"} +{"id": "train_041879", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3e4afc430e82d39f569dbbb536b7ee32fb41026f"} +{"id": "train_041880", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8de3f3861ba4e2d0fcf9b0285768e78927ac7b48"} +{"id": "train_041881", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4aeea3fbe48f37bee20aeeb801fa6a6690f10357"} +{"id": "train_041882", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "400ed9801dd4fdf8bfb90082a6cf760027cb1421"} +{"id": "train_041883", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0f21d9d04825e1dfb6e8e19b1a3c5f4cc7e59777"} +{"id": "train_041884", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0dcd1ceb2b15c39d969e452125bf7a096fa534a1"} +{"id": "train_041885", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0660f3976174cf5359d0b8fb26a02290243ef8aa"} +{"id": "train_041886", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "55aaa09bb88fc26cfcd2f49956c89623538e3a8d"} +{"id": "train_041887", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9fcbb298e9cc84d23c6c93356ad47295ed1b1289"} +{"id": "train_041888", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "59259cc20d7f7e9a5c3472332397eab23d6a38d1"} +{"id": "train_041889", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6092639562db6e52469c274b7771db373139d73b"} +{"id": "train_041890", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fa947277d0603b74a1c127a5598d674882ffe7e7"} +{"id": "train_041891", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "634901acd3fd98a281011d6ad1857cd23012822b"} +{"id": "train_041892", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2a6d7b18017750618262c757519353a5d743ddea"} +{"id": "train_041893", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a5f881287b05bdcd283c8b7b800ae4d39b6b9c51"} +{"id": "train_041894", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d3ae880988a39c1465f861cd753adb03dcf024d5"} +{"id": "train_041895", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "298e2ca2bcbedc4c6c7a855aadc314126e548efd"} +{"id": "train_041896", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "eb5a23b8215206cc4f225e6f7ca55e11c15eda69"} +{"id": "train_041897", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a57ec6cca6327a651182a4c967d6adabcbdfa841"} +{"id": "train_041898", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "945e8dc7e7b6e4b2bc772de6c2b3d2c21939f136"} +{"id": "train_041899", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9f213a78b1898c6654837260184ae81b67be4b57"} +{"id": "train_041900", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f973d2742a5aa5776bbfca657da41655ca00c5fa"} +{"id": "train_041901", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2c0457be206f2799793c379702a8466dce0664d0"} +{"id": "train_041902", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "fac82d338b8dd5b6c6c99dfb1ac265c2d7a80974"} +{"id": "train_041903", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "875f5e5d42edb5ed9a8f01442f24836354989e66"} +{"id": "train_041904", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "23b7a34725d1addbf6510d0460a6d9f4fc500391"} +{"id": "train_041905", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a3ecedc758aaf938aa3791d876a2499f4e5dbdb1"} +{"id": "train_041906", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ef4ee3c849430de8eefbc430653ffdc3d0cd200d"} +{"id": "train_041907", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "eb17cdf700e4579845c08ddfcff59fb5209d17cc"} +{"id": "train_041908", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "898c166ef593033285efb1f9ad1994d249d12fab"} +{"id": "train_041909", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "182a8b8ce57ef02bf7e679ac7bd2266e0ab5e001"} +{"id": "train_041910", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "63ca3536701c444a6eafaeb4439f30c7de58fc9d"} +{"id": "train_041911", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "82d5ab2b744431ccc3f75c7b0e49f888e4f16b01"} +{"id": "train_041912", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ae5cd7b9bb398c1a59138fe7b2c5fec737b66196"} +{"id": "train_041913", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2e03368b4e497eac3e7713c6d2d9a0d17e28c0f6"} +{"id": "train_041914", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2e03368b4e497eac3e7713c6d2d9a0d17e28c0f6"} +{"id": "train_041915", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "eeffedbcd8708460487cffbcab21836bfbb3c2f0"} +{"id": "train_041916", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "07c0344e5297cc64e25cacd3f938345aec70b671"} +{"id": "train_041917", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "34ce41756e37fc6d44b067ea8cbec4af18362124"} +{"id": "train_041918", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3f0fd682929496792bb7768058d9bb95a53b61ef"} +{"id": "train_041919", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ef62059cecc5e08a20e657a36dddf24a7e446fbf"} +{"id": "train_041920", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "64ffa0aacaacb551e7d303cd9bce4a478f278d23"} +{"id": "train_041921", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "cad30b2cca704957706e2035f2734f3ce714a61c"} +{"id": "train_041922", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "842c500881b4bbfc0ca07dc2c7d8cd3db3bf1795"} +{"id": "train_041923", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4b8cce441765b488888b191e53c4f70222ef3685"} +{"id": "train_041924", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "84738828a9af27008f468654ca76ad6e56266454"} +{"id": "train_041925", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c456b488aded9ef671d1704ca7880bf066b828d7"} +{"id": "train_041926", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c58597d7da75faebd1aa9f4a640c2e911667ea4c"} +{"id": "train_041927", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c8b4db0f3d8ba601c4832e18d64a57ab176bf72a"} +{"id": "train_041928", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a1873874d069fc5b7e1a5ef933ccab88b35438f6"} +{"id": "train_041929", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5a6a87bc11367d4dc7c8120c5eeba80889c7b73f"} +{"id": "train_041930", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "02cf08cfa45520ab9a44928eecc6501a30cedccf"} +{"id": "train_041931", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8a04e0f5ed56c80aa8f0dceb7b61458aac9655ec"} +{"id": "train_041932", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "af9fb3c897591526de2fba8eedfd64dcf75479e4"} +{"id": "train_041933", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9bb251b1d848673231e7a52185916918704a0c7a"} +{"id": "train_041934", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d3d4cfc7dadceeb7005a34e381b8f378cfdc5b5c"} +{"id": "train_041935", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "482c105471b1da785b3bce0395bdce56d7edef72"} +{"id": "train_041936", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "74dfab6e55e80c24fa784de70a256617452535ca"} +{"id": "train_041937", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d7ca8155eb72d41d7178783da791f0c96588ad18"} +{"id": "train_041938", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4585a389e099a6a674a8f38f933c2bd6546ee25c"} +{"id": "train_041939", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8e029ead29076d49c02c337852c065ae2278325d"} +{"id": "train_041940", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "03daaf7e52f400ebff8481878c7a5312c956bd75"} +{"id": "train_041941", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e017bf60b7746257a989d00c095753fb1816e589"} +{"id": "train_041942", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6e0b74481a44b0379d791cfb05dc5ea85ef9019e"} +{"id": "train_041943", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3f7df3140f55ec53fc7f2e59e767e60e9e6b70f4"} +{"id": "train_041944", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e02caeb67ee59b9143310a3cef76cbd57234ef98"} +{"id": "train_041945", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "b4886ddddbecf68d1edb118093dc2891986a9267"} +{"id": "train_041946", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ae5cd47533b793bfab2b36013fc5140469e6eba1"} +{"id": "train_041947", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "277189aeea628e93abb491910e2020c1812f4909"} +{"id": "train_041948", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8bd336187170b0a1d926ad0fdf7e6f7d4846e4d8"} +{"id": "train_041949", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "af100504f98161778031f87b4d767671dee40db7"} +{"id": "train_041950", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "7312203d1f0ecd2d3013c0755214d761b4e80f14"} +{"id": "train_041951", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d84c11e6e4217f29ef7cf5c13115b7378d6be221"} +{"id": "train_041952", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "614662976a4d4d8333b340b4b38a5f94ffec3651"} +{"id": "train_041953", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "7b39f187e33b9735b9278872f29687c616b35b70"} +{"id": "train_041954", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3924203cdad3819b6b6f651c5dd61a3c1048479e"} +{"id": "train_041955", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "de13370885974319b5d397fbac6dd8448a3ec9f9"} +{"id": "train_041956", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d43db0585ca2f41fe3483b1036d0fc5e2d4c8d1b"} +{"id": "train_041957", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c57454a4da244067f0c863750e23859003e61124"} +{"id": "train_041958", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "23ff9244e2fa6917f453b9599709b6bd958cacaa"} +{"id": "train_041959", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "63c8ad236fb291c06f34abc363461054a45f4165"} +{"id": "train_041960", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "de6f9ca24521bbf3cac1b0a7fd8aa43ecbfd8e3c"} +{"id": "train_041961", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "4a04bfe9b0f16da6fdfa7c660c777d827ef15eb3"} +{"id": "train_041962", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9b4dafac1bf2bc013420a9ecdcf67e67578d0c4d"} +{"id": "train_041963", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "12dbf7db54a22e0e5d123d159f83937cf9af9e01"} +{"id": "train_041964", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "529a5b86e9fd76f738230b1c66e62557f806f0a5"} +{"id": "train_041965", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d99d9632f7a0611348b28c06f96f97e05e6c89ca"} +{"id": "train_041966", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "244ebfd72762d0f12fc35f61936f2de69e9f8279"} +{"id": "train_041967", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d85a9369eeb988c8e7624334f70cfbaa362697a7"} +{"id": "train_041968", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "dd64063e9de4a02e9873fddf7d00803949da0623"} +{"id": "train_041969", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7dbb125206d44852a14bcdbd4a719747ff639311"} +{"id": "train_041970", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "bcebf0b4e926179bed434593bed12243aae81536"} +{"id": "train_041971", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "38c547184f396e1219360e5e04e16122a1578f54"} +{"id": "train_041972", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "618eea9862917e056eb8bdbc902ed322a2454225"} +{"id": "train_041973", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0beb24f0fa14b70c800c5c8e75f8e3b58af4fa59"} +{"id": "train_041974", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3ca17d4cc877e5d81ee9b07c808cc3b04ce18fbd"} +{"id": "train_041975", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b3da9aacd6d0d36bd99dc9aee8c212c19894e902"} +{"id": "train_041976", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b223380cd98aced7f8093deccc6695638110383e"} +{"id": "train_041977", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "54bbd19f1c95480ec25b66fec41819b4d8da32ae"} +{"id": "train_041978", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2a9f244eba2747497d82d74bc72998102b4d7e73"} +{"id": "train_041979", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "9bf91afb97631df08c03ff339dc3e32d4852f17b"} +{"id": "train_041980", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3ac01de5913705a54369fa4d192383e2b2f2c9d5"} +{"id": "train_041981", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "688c5aa2d7b927fde112358a809e4c3ccdd8ec8e"} +{"id": "train_041982", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "78638b8ed1081a45332eb08ed399912fcfa258c9"} +{"id": "train_041983", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7312203d1f0ecd2d3013c0755214d761b4e80f14"} +{"id": "train_041984", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "71fab7421dc9461c4323c3cc5de6f6da39cad949"} +{"id": "train_041985", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "edde062fe33371071b8ec5fa871ce2cc6d40ae7a"} +{"id": "train_041986", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e94e67705e8145881eedfa888fb70c5ead3c9441"} +{"id": "train_041987", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "875d41342c624808ed5eeb645be194d95f345959"} +{"id": "train_041988", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e0434091a1629283b9e562324e7ed72471e6cee0"} +{"id": "train_041989", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6c134b4f3f5d52ca6165193846cf179107216af4"} +{"id": "train_041990", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "10efeb6562ac55455023e67150b49013198f5636"} +{"id": "train_041991", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3c11277c64d5e34033d5cee4e9424da19158cbe5"} +{"id": "train_041992", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "71da6c734d2e30f90444263a61c826890de6ee82"} +{"id": "train_041993", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "5bd361a0ed5cf1036bc3598be27deec3bcc07332"} +{"id": "train_041994", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b3708c8064e13a002c35dd6ceaae29b7fb4854b4"} +{"id": "train_041995", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c23d55930d6963f172752017bd52edca7c7c0bad"} +{"id": "train_041996", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6e5e7ff2e9659890905f7b541606df327ae57abf"} +{"id": "train_041997", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "954f3976e0babb2d989f9c6d95b3bdec442bdccf"} +{"id": "train_041998", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e9f7a1f5267385da4c4a8daa3aa6678ddd5ecca3"} +{"id": "train_041999", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0305879842c7ecce032cf4deda726f31af43b560"} +{"id": "train_042000", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7adb10256b75668b071b5b6fa554991b1a6ebe89"} +{"id": "train_042001", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6403cc058490676cc72f0c3cbe63d768d74f4c71"} +{"id": "train_042002", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "af05ceed642fc682e6be77a7ed04ae53148bbcec"} +{"id": "train_042003", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8489d2effa9cb7707fd3e1f7bea49f8ff1e36051"} +{"id": "train_042004", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b2cfadb1c247fefecf69d3c9ffdcbdbc259f4ebf"} +{"id": "train_042005", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "91256f7969bb765b60284e8b08710de18f114c7a"} +{"id": "train_042006", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b514d3b2e98e5e07715f3d2f8bc07ab64bf98a6a"} +{"id": "train_042007", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "672ce8e02145200afdfb8b396dc4eeb87b6b10dc"} +{"id": "train_042008", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "43c3e2dbec0d26c38b59f3e1d231bbdcc9efeeb5"} +{"id": "train_042009", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "083e3f72d686f60e0b1437e5f45678503ae39be7"} +{"id": "train_042010", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "213e550f2f4d0d9ce0e2039396dcbc893fd59842"} +{"id": "train_042011", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "30cd7d205e415dd74e83740414b7ba45c74a1648"} +{"id": "train_042012", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "68cdeaf694d3c034901a60c261a14ec2038156e9"} +{"id": "train_042013", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "79d25f3b7d3d61d327eb8ba0d1adc0f77a2e8127"} +{"id": "train_042014", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3f0fd682929496792bb7768058d9bb95a53b61ef"} +{"id": "train_042015", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b9e46799f8a183f17a489612745b0c61d3623b20"} +{"id": "train_042016", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "fe668e7039aeb117a35a61995fe9a758ac98b534"} +{"id": "train_042017", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "00b92b4d12d9b8ac0eed6727f358cfa828cd22a4"} +{"id": "train_042018", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "17cfdaebbae8d33e6e329b385d1aa612c289386e"} +{"id": "train_042019", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "72026cce40f05367b10e14f3b3ddf8b11eca1b0d"} +{"id": "train_042020", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "00b92b4d12d9b8ac0eed6727f358cfa828cd22a4"} +{"id": "train_042021", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "12b5ec1c6aa2b7dc9d2c0db68cf8f0ed5ed2d689"} +{"id": "train_042022", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b6f5615fafbf201c532e2e165a57e08c8adb8cf4"} +{"id": "train_042023", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0f9f96f73ee128c659e24f0e85606c30879b471a"} +{"id": "train_042024", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8baa0bb7546d01bfb36706e14f79551294f63680"} +{"id": "train_042025", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d49213776f545b45bb8e67aceaaf0d88a11fcd70"} +{"id": "train_042026", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "1ce81d5d76c6eae389a61249ff13453bed298af2"} +{"id": "train_042027", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ab5b97d785c4af6715a9b76adbea3b51c9914195"} +{"id": "train_042028", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "317627cfec821b874177ec0b5da586602e49bd9a"} +{"id": "train_042029", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9eb7a802c41451eec5f5518b1411334c4a3fd2fc"} +{"id": "train_042030", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7867592e7130cf42b04102f969e97074c4295cbd"} +{"id": "train_042031", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7330e52f7e460917ad62c25a28b5ad1e8b89daa8"} +{"id": "train_042032", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3468c3744911168c82afd6c20ba9b513976b599d"} +{"id": "train_042033", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d013aebb7c04184991ca5bfbdf5cb500688e2cc8"} +{"id": "train_042034", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "92d063abcb549bfc9e210b192ce7586d5fd03670"} +{"id": "train_042035", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c9979df3daf8d79efa475341e79401469fb7b699"} +{"id": "train_042036", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8a7f6bb7942c9a7188340afa40bed2a0e6cc7136"} +{"id": "train_042037", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a42b138983a384ea4ef9327b4b687732eb2b9440"} +{"id": "train_042038", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a40f712e5f446c5ab15a0630bf21b86bd23715f4"} +{"id": "train_042039", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f4c3be922260f69a87afc15642a858dbe8ad9bbd"} +{"id": "train_042040", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fb1f2bd03cf87b1bcbb389cc31bb5a5a17766c9a"} +{"id": "train_042041", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b97c86d41d1f16ff256b70febe5af01693bc4381"} +{"id": "train_042042", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e3883fac3bb79bf2748baa5e53c720435bc631f4"} +{"id": "train_042043", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a778921611932b713202311ab0254226840a9a8d"} +{"id": "train_042044", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8db32035679914b52d16b4dbf400750c165d229e"} +{"id": "train_042045", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c08deab629be508a1a0388feadc90bfbd3274aae"} +{"id": "train_042046", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f48e35c77c72e2cdea9734b8d5aa3384b3eb2429"} +{"id": "train_042047", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "28ad28fc46c70204e6463100ea47c0666236e4d5"} +{"id": "train_042048", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "6360d3a301d3145c116692b1fd6ed43310cd1d5d"} +{"id": "train_042049", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "214afcb5decaf5c710cf7e246ebfd558c45a7b96"} +{"id": "train_042050", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "6b6440a740ac6f42979654eb66a7d22b3711b853"} +{"id": "train_042051", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b0207863162fce570dada3ffe71e66984da9937d"} +{"id": "train_042052", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3e413bc874760cc9358a29aa347f89847e7a7a0e"} +{"id": "train_042053", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e32e1e7df6a52f62737d8ddd5a9f743bdb73251a"} +{"id": "train_042054", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4b5dbf01714a9bc4304ef8891704c3b432d619c3"} +{"id": "train_042055", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "33479a1755023414213fb0a660e5d7805640cb76"} +{"id": "train_042056", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3e7e741b7cc7e5d12d11cf6e0deed2bcc2a74115"} +{"id": "train_042057", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "684c479b32990eba98d9d0d7ef82f35377988503"} +{"id": "train_042058", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "afb37d24c8a6fc99a9beded795481bb39e63cae2"} +{"id": "train_042059", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1aab72284b4fee0e9b821117d790bd365ec0c683"} +{"id": "train_042060", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "173e439712a4eaafe7fc553c83fde12224219a39"} +{"id": "train_042061", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "24d5d6ce03ed97bbb5def017c37c064cde21f074"} +{"id": "train_042062", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8d859ad647fb2010c54b58d104d0c027280e0324"} +{"id": "train_042063", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0e40ed1a8c0c824241db0b67809a736537f9ebfd"} +{"id": "train_042064", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "bf6332474fc0ab1249d3586f8fe5cc64663493dd"} +{"id": "train_042065", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "e017bf60b7746257a989d00c095753fb1816e589"} +{"id": "train_042066", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "79f985e3f3aca6ed482a17d1e0c4852bcabc9068"} +{"id": "train_042067", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "27f5869b05117b8657bd7591fbdcd858166f7b18"} +{"id": "train_042068", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "cffed426d93cd0c9de5bbfd78b564009b2cad21c"} +{"id": "train_042069", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3e5a534108e488bbb0264c63425ae802a39c1441"} +{"id": "train_042070", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "51c6b0a3f36ee5ffd441fd9123f51a0a33da46f0"} +{"id": "train_042071", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e9a5e2bc7029e9aadaeade7d45929bc4380c2100"} +{"id": "train_042072", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "81fc2ca908ea801fdc2abb562ff29f81e2987c6d"} +{"id": "train_042073", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ac5709f8113c533fad4e6c749acdc2d5e9bc5ab6"} +{"id": "train_042074", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b35dbbc5b517fb701256e0d7fb62409a8f8f4f96"} +{"id": "train_042075", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "1fe9c7616c65cbdca95db74a01e3677bdfb4c2cd"} +{"id": "train_042076", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8b1d06b13f795f77e8cd43e80d823e410d727d6b"} +{"id": "train_042077", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "82848572e0099984e3e26c77f26e773ceb588d40"} +{"id": "train_042078", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d0e4f7885824c5f666b8e2d52753ca12111d8368"} +{"id": "train_042079", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "1de91338ad774d294bf86ca0502f30a259bf1117"} +{"id": "train_042080", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "72715c06cdb4b1f0eb37da167bd3e2f1d400e578"} +{"id": "train_042081", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "705e8c5dab71f7fb4ee57a46492accd411cdfc1c"} +{"id": "train_042082", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2f4ceebd2697b921b04d334d57ada19e9de229f6"} +{"id": "train_042083", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8a180488f06b7585aafb526c3691da2d7a24f7b5"} +{"id": "train_042084", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6abdb25ba8a0375b063cd3306780fc267855bc65"} +{"id": "train_042085", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "00002db1ac8f904315b39eba6376ba8cb73bba2e"} +{"id": "train_042086", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1bb2878ba6f9cfc4332fb009ac08c2d92bcaadb8"} +{"id": "train_042087", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "dc45df23b49c682bd8d8b03bf3e37e08ff534653"} +{"id": "train_042088", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f2e44ab71842b38e916a1de7c1b6d736f19b6245"} +{"id": "train_042089", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "acc12ec69f7f6e28b6056bcf1f94f7bf2d649a70"} +{"id": "train_042090", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "eae673d632652a0eafa9d33e197fa83f17a1e981"} +{"id": "train_042091", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "dc7980699929340416d490334df26a8227fab3d1"} +{"id": "train_042092", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bc6672aafbff49700d6b64a4bdd37c41b4d31075"} +{"id": "train_042093", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "5243a52d1f9c186b5639608891b205ea7f9b7d45"} +{"id": "train_042094", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f91d3e213a17c8314a3a71463b3fcacb9f8e45d7"} +{"id": "train_042095", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "16368252d4a828d8853b04b31707a67071dd167d"} +{"id": "train_042096", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "fe29695897cfc9a19a81cc1794ce1ab7b8d3026d"} +{"id": "train_042097", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5708631620e2e0c4f7b72d04528de7f725aeb545"} +{"id": "train_042098", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c930d1e8383e7b4f1d51501e223fafd88c7521b3"} +{"id": "train_042099", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "57adf6a11f29a3aa620daa7ff32d4622d3ee0313"} +{"id": "train_042100", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8526d955e0f77a9d8a64108537f0e0e33fdb80c1"} +{"id": "train_042101", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "bd7a34d47c94cd17cc59fbad338c0cb514860607"} +{"id": "train_042102", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "c1d1c494ed6d10b0750e0ad9d554a2d867e62968"} +{"id": "train_042103", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "7119422ca5a83286d4ac6e192d1a7138719b42b4"} +{"id": "train_042104", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1505bc9c477b6436533d778be7ea8a28e68f0a1f"} +{"id": "train_042105", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5ebeac9f98cb0e3cbb60963cab03e00bd4dfacd0"} +{"id": "train_042106", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0f69d8be694aa09911c9e20a4609d6a39955370a"} +{"id": "train_042107", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "20b95cad86311b865d68b9fb39b0dd9c985a05f7"} +{"id": "train_042108", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b3c8a52036eea13ef1c455e62a9f294a7640eddf"} +{"id": "train_042109", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "cd36cc580bc39412a29737c4e8ac547b085255f1"} +{"id": "train_042110", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "072ec5350469d505b4c50a8393065af766cf783b"} +{"id": "train_042111", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "de60a364d6e6dc7def2621e8d1729b3d846fe430"} +{"id": "train_042112", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0be5cb9e0e770906956188efb1561b4c7e0f6ae1"} +{"id": "train_042113", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3042a6e82ef2325163d55f2bba475f3f11d8a3c3"} +{"id": "train_042114", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d36a1a9b6002de444050a66a5e1679210d0d15a2"} +{"id": "train_042115", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f77160ac1fced445cb25d12a22a30d6975524db0"} +{"id": "train_042116", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b1d0ba0fe10fee4b9cea11f07c309a0cf2ba222a"} +{"id": "train_042117", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "32c9a0173832845415f3560c8ba6d66601663b01"} +{"id": "train_042118", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1d6a0357c276e1fb2e693675c73a1e204cbf2d4c"} +{"id": "train_042119", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f6423adcd70bb7f5406407f3e9e6cee39f3b929c"} +{"id": "train_042120", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5438d98ec27ff43a6af5907c6c31ecad162a94ea"} +{"id": "train_042121", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "bde45e34bc87ff0764cb951b2a7897faecff1d28"} +{"id": "train_042122", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "7bb83327af74288dab89b84a1896ef6e152da571"} +{"id": "train_042123", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9a98b725dd69eee3d514ebbf55cdfa4e9afd19ea"} +{"id": "train_042124", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "89934cd3df295002cbcfa81ce2df5b29f13d13de"} +{"id": "train_042125", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "56f0ad3d38ede9b0acca3cf0d18ce9f4a7edeb29"} +{"id": "train_042126", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c07d43d5e3d7eb1d288377b13af776dab2478cd0"} +{"id": "train_042127", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9275df2864457178aa34eada92f189977e810364"} +{"id": "train_042128", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0893e2beeb08a9785ee78602535b6e018921bb0f"} +{"id": "train_042129", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e8db5bbaaa8b1468d11e72fcd122614d8ff047b6"} +{"id": "train_042130", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "15dcc04785feb6229b3f773c523574f0498aee6e"} +{"id": "train_042131", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "317627cfec821b874177ec0b5da586602e49bd9a"} +{"id": "train_042132", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "7189727c8053d4104c9ad481428b509b891a901a"} +{"id": "train_042133", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f7de22fd3c087950e422b0e652606c9b94f8baa8"} +{"id": "train_042134", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "b6776d5d30949dfe7b1afb9aa81dc763b012f91d"} +{"id": "train_042135", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ead4456c9f322c59a9b23ac693a6de2be55ec272"} +{"id": "train_042136", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "4abdbd5eaed653dde60ccd8a5de1bce2a188f706"} +{"id": "train_042137", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4ab057e5ff30e5b9faf743eded7eea42c88c71f2"} +{"id": "train_042138", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "87785d212ff2d46fee7308e60f09ff2e2802a25f"} +{"id": "train_042139", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a2654445a351814498bae4134a96e04e19bbeda5"} +{"id": "train_042140", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3f9654699ef176d60d01741c33206ea9ac3163ff"} +{"id": "train_042141", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e9dd259f5df1ad71956b7418facb4e1ead214433"} +{"id": "train_042142", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "83a68b46bc5d72038d27016d35361e78636764c0"} +{"id": "train_042143", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "9fa5f3349f429a15e4b403988470fe0ae4fc5524"} +{"id": "train_042144", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5ad98cacf7678b3ebdbd18e8c55ea6a5f211e550"} +{"id": "train_042145", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "fe44ed408fd7913c3895fec1c00be4b0f1100fe0"} +{"id": "train_042146", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ec12812daac6d68b3f2beb51ad2233c8e78705d8"} +{"id": "train_042147", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b5ba44862c5a40ea16f0bd14f106ea3e07bee0ab"} +{"id": "train_042148", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "cda71665b97a043dce1c96c391e5ce005e93a45c"} +{"id": "train_042149", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5fdbd84f5662cf4f43722caf9cc782cec7eb3c35"} +{"id": "train_042150", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "65ffbf2208afe9e075075d7b33fc7bb504d85644"} +{"id": "train_042151", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "fd44504e6689bf40ced3415a2f263ed774818be0"} +{"id": "train_042152", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4e17b9af519c240601a94c0e16a7f9b2526044b0"} +{"id": "train_042153", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "168fd0f33ee8484261c9254ffd0b2e21ea6ef6a4"} +{"id": "train_042154", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "838259b8137ad6b2ad616f2e529f2f52f9b836b3"} +{"id": "train_042155", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8e103843d5380983a087c5dc76efcf7c71b7b748"} +{"id": "train_042156", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6faae427fa5d212d9dde900bbd4826e48d02ece4"} +{"id": "train_042157", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a4e1b2f23447622ad79229f1f68e764a0ebdd93a"} +{"id": "train_042158", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e789d8b38ac74c09676d61425d3967a853f2b5a3"} +{"id": "train_042159", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c04a274bec4d1eea6a73755e99c496c43f2893c3"} +{"id": "train_042160", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "81e2b0951190ec9096db73fab973918e701547ea"} +{"id": "train_042161", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "6c50a9daf8e3f849362d13450a5b87be6cbb59b1"} +{"id": "train_042162", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f990c3e314da635b2ba55a8fe67188ba3db194e8"} +{"id": "train_042163", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "16c8676f995e85262729ba425e25676ef279223a"} +{"id": "train_042164", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "691528075709cc387c08507dc3f6571dd04efb30"} +{"id": "train_042165", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "2b3c52a422e233acd9ef2f41e7271b728c87aff8"} +{"id": "train_042166", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "33485a071d07bc780c6704efe6bcea6183ad80db"} +{"id": "train_042167", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2f2650310de2378a47a360bf7b6872e756c2af9b"} +{"id": "train_042168", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7920f9091e787a78452df30eacb16291244796a2"} +{"id": "train_042169", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f4ab167aa0859a504c6369d8a012a9d3e2898de5"} +{"id": "train_042170", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "16698a92495930f4cc4b8ee0d076d18ef451ea5e"} +{"id": "train_042171", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "fdc73edf8210b9aa37dda4f24caee0379cc5bc64"} +{"id": "train_042172", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "aa6c26611918924a00c7d427e89fc3c53d62806e"} +{"id": "train_042173", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "33c5b6dd3286dc7838e74bc2c04ac7bd6fbccf31"} +{"id": "train_042174", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "545eba322603c9e5cf50876574a46453c6f1d68d"} +{"id": "train_042175", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "673a5c3a70edde8c972b7d233dfd399d3349f0e6"} +{"id": "train_042176", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "81a48c1bf2d9c068f47104dd98f58f9f0730e1c5"} +{"id": "train_042177", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "66d07327fed08d3f715664d6a17a139620295a06"} +{"id": "train_042178", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7b85afb6f163821545a158023bf8f71f96ad06ba"} +{"id": "train_042179", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "79decebb4a760ef1f4041dd206ef9cc45a696b2a"} +{"id": "train_042180", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "cc14326abeccc55ec91c499b7f3abe1bfed5bd2e"} +{"id": "train_042181", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b189897a0811f125d584898eb34ee0c407783d98"} +{"id": "train_042182", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2e2755c2b1958b8bb6fb317b73b4de5017ae3a11"} +{"id": "train_042183", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6970355ba93f92c4928840b6fa9d5ce28cc0eb0d"} +{"id": "train_042184", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2d768d959a4338bf2071f61935fda646a83b1dc9"} +{"id": "train_042185", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "00f6b72a17fd711dd23ec429e0ae5244348400e4"} +{"id": "train_042186", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "de26e42a42e9b4e6a31868fc4d66a142c5950f68"} +{"id": "train_042187", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "cef669ed1e326fb72d3d7b292fd5e3be89f57570"} +{"id": "train_042188", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6f02359e139ef17e66177b22350b8332c27c2396"} +{"id": "train_042189", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ce6caf0ad86798d0e58db653d1f4e21d96801f90"} +{"id": "train_042190", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "3c80a6ddde80a6840c3db982c7dba20edd94fb6f"} +{"id": "train_042191", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a5bbcd1db99d4606a8dcede1f71d918082e42bae"} +{"id": "train_042192", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7a3b30c7d5c07e8cdee86033d80e5a5d89017e5c"} +{"id": "train_042193", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "6c88bcbb846ba50779200231ae98f1b41d6ff450"} +{"id": "train_042194", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "78b8cde330ceda46c906fba92a5def3b1f4fe7af"} +{"id": "train_042195", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "15e46fdf0c1fc333a2995775f3fb4dbf4bbb1ca1"} +{"id": "train_042196", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "608055982bf400f0a2a71ebd49e8313255606acc"} +{"id": "train_042197", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c418adbe68abc65b5170adf9bfedcbbbd23b05b4"} +{"id": "train_042198", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "6fb29b368e1db895dd8abd372ccbe039d3e5613b"} +{"id": "train_042199", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "18afabef105d2e09dc848349a04c8eb02538e3bb"} +{"id": "train_042200", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "1ae22066acac5569580f4ce7fa02900ab8559e1a"} +{"id": "train_042201", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c228268d55ae894870754688b3293a2ffb5f4b55"} +{"id": "train_042202", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "300521913debacb33ab6fe66b11ab04b64589d3d"} +{"id": "train_042203", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "af800bcc9d57cefc93c725f95d44571b741a9105"} +{"id": "train_042204", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "134ea907f0a52f228044d04ef1c94687131bdada"} +{"id": "train_042205", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "bbaebca6e46169ae8afaa1466d3be1577056cf7b"} +{"id": "train_042206", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "af0df436744dbee227025602be0c3fc3047a1af1"} +{"id": "train_042207", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "25e610d39e4d2070f216b75f69fe87d9d00347fd"} +{"id": "train_042208", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "59a5f0a035a75990af00535dcf976da513eca136"} +{"id": "train_042209", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0b248d66f7d8b64fec02cf2913e8a8a269d4a3d6"} +{"id": "train_042210", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b25439f94471459472a48851975e7b0a9887362e"} +{"id": "train_042211", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c238b7004ef99bbeec1deca6a788bbd8a541d8c3"} +{"id": "train_042212", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6ab65e5bb97549afbf5a629d08dc25916f9d34db"} +{"id": "train_042213", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8f261c80109e98f0a89799f189fea8ab0581a22d"} +{"id": "train_042214", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "432004cb124b06041b52c18cff0d7ea57d38b578"} +{"id": "train_042215", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f8ad6e3711c251103d6e6d686297583e84eb1e49"} +{"id": "train_042216", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ab1fdaedddad235fa49f1d0d0e016adfc3be3c40"} +{"id": "train_042217", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6a1f68ff7cfab7c47439ca747f1ddb06d12376a5"} +{"id": "train_042218", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e6291762a6c5e02f22b0347ee663ba6089daf0b6"} +{"id": "train_042219", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "73a107be8281f9735011314e89401845530c6fad"} +{"id": "train_042220", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0238c469819413189826a755bd24eb0b2c984a3c"} +{"id": "train_042221", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2284b36d4bdb150273498300c74f73ea4539f71b"} +{"id": "train_042222", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e9943bedea11c8d5922edff90d986ae3afede6eb"} +{"id": "train_042223", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "8b0d1d56e58a5f7c30d923aca16675c2aa029003"} +{"id": "train_042224", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "31c82ffb73e5baeb656bd3d763b754b6d004bc3b"} +{"id": "train_042225", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a81fdb7cbb37d53cf816597781881405118d5b33"} +{"id": "train_042226", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "42bd266260b033157c1aa179b18e5774f7b8639b"} +{"id": "train_042227", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "1d69de07268ecb130fae928fc921c62d4e92ac62"} +{"id": "train_042228", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "797f524e04a12fa54749ab4117b6ec5f38b19ca4"} +{"id": "train_042229", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "bc961c94d0bd298a0ec0b529df5a37d6396e816f"} +{"id": "train_042230", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "61c2fe71b4f2367ff1cb3a988167f484d46acb88"} +{"id": "train_042231", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "817beb09562077595335fe2347dbbdf21aa1191d"} +{"id": "train_042232", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "34201a250b9c4de3b5c2273b3e167c3b1fe63c6c"} +{"id": "train_042233", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c4dfbe86124ee357d51a684c77d4ef4d35d70171"} +{"id": "train_042234", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5f4b81579332f5adf6c5a0a8080800bcec0a537c"} +{"id": "train_042235", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "12179ded5654cec9cfc678d325a7a941441b5cc1"} +{"id": "train_042236", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c01e2a88423d8c5860ee456e8e64d5bb98c6d3e9"} +{"id": "train_042237", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "73c0cb3fe03d0357d3dcb6ed5458abe89f54cc81"} +{"id": "train_042238", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "805b9711f10fca2cd9b8c5e38c41227bcca3cdbd"} +{"id": "train_042239", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c26fce838262a39c6fd4da4504f0efbf76db53be"} +{"id": "train_042240", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "fdc73edf8210b9aa37dda4f24caee0379cc5bc64"} +{"id": "train_042241", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "844f0f5c0183b301ad0dd8ab3f9aca5e4c165c44"} +{"id": "train_042242", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7afb90869ce272743f6328823f1161c087ffa274"} +{"id": "train_042243", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b0a7468bab2a44a946caffa3f788062439368250"} +{"id": "train_042244", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e44b50090d7a38d2d6b013c78646539f23623b83"} +{"id": "train_042245", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "666949d029a91d5c0912fa5619e4535a4121ab0f"} +{"id": "train_042246", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "32eae54258de61f8a51795c2acf742934995d853"} +{"id": "train_042247", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "61df814bfa99ae0fb45c4b7481912d1a28cbb06b"} +{"id": "train_042248", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "19e915d30bef0aff5e28bde3bae05135db8b8d11"} +{"id": "train_042249", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0464e0d3289ff91a2c31fdd5d69388b511f03095"} +{"id": "train_042250", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "d7c57de796694f28b89ca762aeedb9cd321759c7"} +{"id": "train_042251", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "cbcec364ed384ced79028c2239e7e2a957d020e3"} +{"id": "train_042252", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "916f8895aa83c5ebcfed036411ddc8af76ccfe20"} +{"id": "train_042253", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "91aa7fe284bfd222eb9392398b42769c36592f9b"} +{"id": "train_042254", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2104eeee33598c56abc3b3b2284e8298e90e8a3c"} +{"id": "train_042255", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0ec0e424d2b2728e508e76c9c5f12720c40cd150"} +{"id": "train_042256", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "05c440609fe5fd6001cde0bcf4aa747ba13e4ce4"} +{"id": "train_042257", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9638f31b1d86355deab19dd28e80adaef5a475bd"} +{"id": "train_042258", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "34bb83c7e6f2c40f5d637ed93c2f9f0e38cd5151"} +{"id": "train_042259", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9916cbd38ab1ebe6e30e69f43370c750ede438a7"} +{"id": "train_042260", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "049bc3ad2fe5bed8f10acc973603dda6f57b80f6"} +{"id": "train_042261", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "310d82f64b537f42426ff181fa5a4331d7f1434a"} +{"id": "train_042262", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c5c36c74fbd39d336e27b719233ecdc1d0da46bf"} +{"id": "train_042263", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5f841cf58f591ed88247b70fdcddb68f220b22f2"} +{"id": "train_042264", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "85c96aac52e2231823a3eb5b7db4673b0427a8d8"} +{"id": "train_042265", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "95692c01b90bdbf0daa65cedd49ad8abde0491f7"} +{"id": "train_042266", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "fb645a70c140e21e5bb772aedaf8db715652ff20"} +{"id": "train_042267", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "89dd57e462d0d7368216922b2ffdb745a5af6987"} +{"id": "train_042268", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5084c1fd1b21ac11e06f3e829158f0e3682b6db9"} +{"id": "train_042269", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4301d140bf2a46d07ff0b5396a166e02a0e77d1c"} +{"id": "train_042270", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "af0fa4c8476142ff995a456569d25dae6eb28129"} +{"id": "train_042271", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "5dfb467db2e487566689fcf97b57638146e5912e"} +{"id": "train_042272", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "39dfe2411dd153cfc8b89ceea439d6c7e793df30"} +{"id": "train_042273", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "7007da4baba9481c1c44cf68a37b63f4f5a8d2a0"} +{"id": "train_042274", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9b6fb6b64c09c0753b43378c2643e8ef9e1c323b"} +{"id": "train_042275", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "47fa557f82bb6dc6c70752851f14bd7ee13cd391"} +{"id": "train_042276", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4cd76a0f88f9fff26ceb92fe563eabefa29fa109"} +{"id": "train_042277", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e88f043bd4a79e903965d4463799fa64b16f9147"} +{"id": "train_042278", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "5ed76217140e65d3c494f24fc7f96dfa105220b6"} +{"id": "train_042279", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "32758860cdfad6c15ff250df079e271f389ca2bc"} +{"id": "train_042280", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "6d17c50e66babffb8f9958013db2f2953477d52e"} +{"id": "train_042281", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5b5d87c9ac72534e2ac44ee27d648f9993fcf4e4"} +{"id": "train_042282", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d0bc1619a8c5e2c5e1d168bb2265e609b0566b24"} +{"id": "train_042283", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b8713e9b9ebbad100f01d4362402d92976f5f27b"} +{"id": "train_042284", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1230e76b7f4704965b3e9dd6d666990d0a81a8f0"} +{"id": "train_042285", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ed8f398b5053bd80e1b2b585c936f3bd49c63402"} +{"id": "train_042286", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "bbe8b007a8337940404bef5cd5e574040a64b49c"} +{"id": "train_042287", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1bd3ea4e8192ad88e074d60e7c3beaf0fd3ce347"} +{"id": "train_042288", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5634594a5900bf7bdebad0884bf05041cd886f48"} +{"id": "train_042289", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "987e2e0a1caf514ad0972523a60dbace71a3aa2e"} +{"id": "train_042290", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "94064bedd82711ef4fd31036a2061aff09c65c0d"} +{"id": "train_042291", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "cd644527648db31004ce1cba5067dc2e36b6ff8e"} +{"id": "train_042292", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6c415f1855a08a03391579d4b904c4fb5a3d51fe"} +{"id": "train_042293", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0fe5ea2bfd49d0f264fdc3492d5e24cc077ee5ff"} +{"id": "train_042294", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "630ec416b219c9b336567ddd4646bf88936a65ea"} +{"id": "train_042295", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a6655417db89d71e8f00a55cb0cf614146c411bc"} +{"id": "train_042296", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f6803572862187930deb3bc0b1b67c73abaeb9e8"} +{"id": "train_042297", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4825df6d9557c2f7fdc73a5f4335144f7248d451"} +{"id": "train_042298", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "5aaf5d6f63e739466de648b21b6b9b078fdabcf3"} +{"id": "train_042299", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a99301e9683895b92529c453a0df2d5a5cb721a2"} +{"id": "train_042300", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "61394271a259bc80384d0355e0c97c0f8ffe7230"} +{"id": "train_042301", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "60dfb181e06eca47679a5c0f455dc9bec90fa25a"} +{"id": "train_042302", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e9943bedea11c8d5922edff90d986ae3afede6eb"} +{"id": "train_042303", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1da212b11863caf21e0570136941a4c7e749d297"} +{"id": "train_042304", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a2b04111458c333ad8c1bd8c3921819a93f36495"} +{"id": "train_042305", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c1cad9ac1d09a4f065ae4c02cf48e3e03c8cfcc5"} +{"id": "train_042306", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "13447e93ae0b693d2d81a2959f400b6b85a22f5c"} +{"id": "train_042307", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "794f72b6d022e8658a7b99b638a99108eba91948"} +{"id": "train_042308", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ea3f60cfa0876478e42f8d9bba85eb24b905efcb"} +{"id": "train_042309", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b76ae652fcee4aafa1ad1b7c95650c61ce81c6a9"} +{"id": "train_042310", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ff3d87ef688ea2e616ff1300bd6e332f55e7faca"} +{"id": "train_042311", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "883ea5512d802561041338a674fa82d1ba56da22"} +{"id": "train_042312", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "1b5fa9363d4b1fd92085926d6f798ec8a7005cd4"} +{"id": "train_042313", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e4d61cf3bec1927ec9b1699253445078600c08fa"} +{"id": "train_042314", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "01c6304ccc1aeecaaf5a4c83e288651614ff38dc"} +{"id": "train_042315", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "1f4efd3ab0ec4e255ad9011f2663c3f4067a516c"} +{"id": "train_042316", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "48162c789d7427adfe1ee66f58b7ce7815f9c22c"} +{"id": "train_042317", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "81952a136a9374fd27807aec9d0ef3902309199a"} +{"id": "train_042318", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c9d27069241e536bbc1ac6754dc0d3ec74d8ce4a"} +{"id": "train_042319", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "965ee575acd2c8994b7cbdec9387b0eedef62285"} +{"id": "train_042320", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "001b9390b2c3d857be4fca63ef9e297d4298e32b"} +{"id": "train_042321", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a2b04111458c333ad8c1bd8c3921819a93f36495"} +{"id": "train_042322", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6f011b2e04fbf6281040b09b4cd3af5614bf77c3"} +{"id": "train_042323", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "bce755450a5f875a84b6332d99b08337d6864827"} +{"id": "train_042324", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f4ea7ad165742c25bd6bc17975806331fb231c7e"} +{"id": "train_042325", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "cfd930c8b3bee274c30affb9344167334d839520"} +{"id": "train_042326", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "cec3a677b37e9979b9018ddf99f504443c6f6216"} +{"id": "train_042327", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b89f02ac3e4a097b31f88f85d9185cc863cbe7b6"} +{"id": "train_042328", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "61e0f006addd3e7c95b03579370955cb535ec909"} +{"id": "train_042329", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "39349d4a8c5fb334fe0ca9f56dfe7b9ab8e682a5"} +{"id": "train_042330", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a0deeb729c4589546c2b08c989917d5fcd979391"} +{"id": "train_042331", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f28e71735e730c0883a0e41f491c6442326bb766"} +{"id": "train_042332", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7f56e093ebbca71cb9f01c483843756c9900aabf"} +{"id": "train_042333", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7ab3315035a7d411958e94c59179431b8f1e39ee"} +{"id": "train_042334", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e927d7a23de86170a7b23b38114e04ab50d8a988"} +{"id": "train_042335", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "efc5656f247c0c828fd48d9d65243c0b8f1fade2"} +{"id": "train_042336", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d2016e65802d922b8a06407c1a1588e29490f4a0"} +{"id": "train_042337", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "cce43c5c22f5db098d8aad85751789e097f2ee90"} +{"id": "train_042338", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "71c706e8cc2160fbe2856f3b75d4950e259ccfda"} +{"id": "train_042339", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "30329bb94dcd288386ddbf1154c434b0ef4ea9b5"} +{"id": "train_042340", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d3e4cc07faa7ebd15eadc9225af6ecc8200e9bf1"} +{"id": "train_042341", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "678ae6324e32b2ea92e1005605a87384f785fa4e"} +{"id": "train_042342", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "bdb8f3e3a4f763b8d85ff5a0bd16ac1f622dd390"} +{"id": "train_042343", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "df52cfb54d55529430fb514ba61d5c815271b746"} +{"id": "train_042344", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "1507a709d1c4233dd981087325e10b0d21e4e8c1"} +{"id": "train_042345", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5f59ffd5f97e120a8abd8bee45372f7efc19d299"} +{"id": "train_042346", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "4c5acb5ca3f99e7db0e5371fdf2bd657e27a0da8"} +{"id": "train_042347", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "388ea4cdfa916291428fcb2013c9d93c7d5e6810"} +{"id": "train_042348", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "171579cfce59abe5786fbd7e983226ddf647a9c6"} +{"id": "train_042349", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d28074892ab4fb82a3bbedcf99ea2ed9b148abdd"} +{"id": "train_042350", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3caabb52bfc7f1cbec328085cd89f3da16ebe1de"} +{"id": "train_042351", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "bdd778786ee22fe0abb659a4c062ef9091d1a1d3"} +{"id": "train_042352", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d784f9f8f448c3df2204c3340e0be1167e5e84b6"} +{"id": "train_042353", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "5d11d822c01862345a6709a94a20b25ff3693375"} +{"id": "train_042354", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e6afcdab9cfe43c5dc419036880bce0fdcf881f2"} +{"id": "train_042355", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3cce7e9826ce49e827b1c3bca2ded1a4f7a58770"} +{"id": "train_042356", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4e6b2197ef70af333afcf18dcf284f6f100ea01d"} +{"id": "train_042357", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "429626b4a4ffd06edef6779e7ab2cc272981015d"} +{"id": "train_042358", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "58ac95101aab8726c77bd68e4b0a3fe76a12fbe0"} +{"id": "train_042359", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "234058a11fd0fdb1a307bc8c0f3995c8fa775b91"} +{"id": "train_042360", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a62b270e25268ebc9fe750cafc38886931d73195"} +{"id": "train_042361", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c154ebddbfa1718cf92abcbfe844d0c6984fd202"} +{"id": "train_042362", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4d92699185a0cd76fc342b4600fc1ca7f6d2c79d"} +{"id": "train_042363", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "18101423af537a879cb5a3d1ac4fd5eee2d6d436"} +{"id": "train_042364", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8c2734c1f93bd4be8648f54d0bda9ea4f90d0068"} +{"id": "train_042365", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "4b1311b2098a93103b0a52910d88678da12c34e2"} +{"id": "train_042366", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "49672075a5085c49c49d277f565b0172dede9627"} +{"id": "train_042367", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "498a9e7f483ba10db79fb3b6582e9b9be3c5864c"} +{"id": "train_042368", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a998bc8b814cffc83331a370f4e3dc3ecab7ccb2"} +{"id": "train_042369", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "cd561c252f5e9b9006bbccd8f89aa231f12e1a9f"} +{"id": "train_042370", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "83b4d918c838f2ee8ee5cad82795f5613781a5b4"} +{"id": "train_042371", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "873647e86958ffb2dc68155365f8a3efa20f42c8"} +{"id": "train_042372", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "582c6580e46e05f6705ca1704fd016ad7ea90fc5"} +{"id": "train_042373", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c2b6efefe1e24aab94f959f966f306c57971a661"} +{"id": "train_042374", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0160eaef4c0ccffd059114ac46c348bfc27ef156"} +{"id": "train_042375", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "30e8eb68b817e10c1dbfc4f1354c3aeebaa274fb"} +{"id": "train_042376", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8191d7b50649fe09ca44c673476fac0ac9414fbe"} +{"id": "train_042377", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7f791ac4684c1092d0fc6365c79d0f4d97ec6de8"} +{"id": "train_042378", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "01c3ce1e07faa4fb9c3d1603ec4edc2f06371b6f"} +{"id": "train_042379", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "5fbcfd5d4ef79cfbbd52c05e9fe716920acf2392"} +{"id": "train_042380", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "39fd641b58200730740c649fd6e617f9b3272315"} +{"id": "train_042381", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f5dbb675c062aae94045c852a386ffd55b78c3cc"} +{"id": "train_042382", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7d8c952ee8239e846c43fc3751c4bce60a092580"} +{"id": "train_042383", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7e25322ad52439477833d62f2e503dea9f2210af"} +{"id": "train_042384", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "8491ed2f2328fe17f3ab5af6e1e2e2fa2322f109"} +{"id": "train_042385", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9ea75eb9497f6ae5eacb3077ceb40e7514c73b0e"} +{"id": "train_042386", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d6adc26e5c03dfa6e9bdce5be380d462b1e00c19"} +{"id": "train_042387", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c4440c5096ae8b7d93ade5d35955f9865f456362"} +{"id": "train_042388", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "6d48f9391d5823d521b5d254cb72b0f2e55b884d"} +{"id": "train_042389", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "36adf1f8b1899bfc4e49d6d9f26aed8f58a06e16"} +{"id": "train_042390", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c4457d9efe5e282ae3f5b23f454b69b72eb79543"} +{"id": "train_042391", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8bd336187170b0a1d926ad0fdf7e6f7d4846e4d8"} +{"id": "train_042392", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5259f27968afe07c1d95d99ebcc37a234134f9ef"} +{"id": "train_042393", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ca3c203bcb021f1b99acb9058868b4e20912f91d"} +{"id": "train_042394", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0f4b7f8c0daacc1d991863b1f8feceaaded2b099"} +{"id": "train_042395", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8dcaec8af20436e2bde32490b89197f2c80added"} +{"id": "train_042396", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b7744c10fefc61ebced7dd6665236fd1afde9f0c"} +{"id": "train_042397", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "356243d1b5b8dc44729554a59571a68b128445f1"} +{"id": "train_042398", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8961118b61d6781ccde48162790ae04671879ee1"} +{"id": "train_042399", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "1029e6b7b6015a72113d3e552368621f23a48c22"} +{"id": "train_042400", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "bdd5c84d523f32aa3428375d74b51cdf7498eee1"} +{"id": "train_042401", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "65dee2e27ad149848abb25520d93fc976a9ed7d0"} +{"id": "train_042402", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f7aebd0734651b5e6151bea02adb53bee331c869"} +{"id": "train_042403", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2e6b5cb32e79649cca90bf808654d00179e689d3"} +{"id": "train_042404", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "22336b076bc207d0d419ab48759c3e387e26ae4b"} +{"id": "train_042405", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6b6440a740ac6f42979654eb66a7d22b3711b853"} +{"id": "train_042406", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5a84e258797742bb2401ffccb7dcbb50862998b4"} +{"id": "train_042407", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "cff6c69d8da0f579559f01cc675038b958fb58f5"} +{"id": "train_042408", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f5dbb675c062aae94045c852a386ffd55b78c3cc"} +{"id": "train_042409", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f03834f21357e481cd8d290c8d91d4b9aaf2a1f1"} +{"id": "train_042410", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2873e210e52e3b91ee1e878a255bbe67fe937432"} +{"id": "train_042411", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "cc8f9326672967a9272e098eef6af4acde44e738"} +{"id": "train_042412", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6ed063afb80d0da179d974c628701f67b2f91836"} +{"id": "train_042413", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0039a6f98bb725bbc0f0c073e8ca8c742f3f02b6"} +{"id": "train_042414", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e9a26e37bbc6cf5beb7e0d7f81264f32e5361e46"} +{"id": "train_042415", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ba24666a7c93bd9b6d3434ab68a30be45688d1ab"} +{"id": "train_042416", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "8891bf3ec0bad94b89abf498c628208ef321b98b"} +{"id": "train_042417", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "16f24ff1a13c787a9a48ca5b122178fc33ea5d72"} +{"id": "train_042418", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "71492e4198a87e4931939add80f00163b79740ad"} +{"id": "train_042419", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9b20fa003a6e7c51c6689e9548b4179ad8451c6c"} +{"id": "train_042420", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f818f8dcdd682813b0bf683c4d44e957122d97db"} +{"id": "train_042421", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a5f749c4e4b75527b90eff118a9a3913caca331d"} +{"id": "train_042422", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "00f6b72a17fd711dd23ec429e0ae5244348400e4"} +{"id": "train_042423", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "340a9ed76da815fdcd5b2960d9e086c7af8f49c0"} +{"id": "train_042424", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "64264f4d3dddb49a4ab71d62305a831325df3031"} +{"id": "train_042425", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f07b9d4f5ba14b3f09ca8e7b7acc3fca7be40362"} +{"id": "train_042426", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e539a837717c3a39c07fa6dd09f1f74ca706d850"} +{"id": "train_042427", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f09073ac12594d4b36181462019b52998c0d9a70"} +{"id": "train_042428", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8e8b6f0846b1ecabb12da717dd710b79516254bd"} +{"id": "train_042429", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "dc4bdfd6e750abde08ce8cb481feea5178a5992d"} +{"id": "train_042430", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "df67328c845b8f104798c031fd489b6cbb8952a8"} +{"id": "train_042431", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e83a38ed3aabed5b2a3f673fa35c9206ea390f53"} +{"id": "train_042432", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a37eda737c0ce6a69db0b4cee11dc7a0c5d8fdb0"} +{"id": "train_042433", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "37f0d76722e455269837bdbfee9fed2635b95251"} +{"id": "train_042434", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4f7d6f73abbf496ac9a4b61fd40a88bec8169e7f"} +{"id": "train_042435", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "33c238c4be49a3e301c5a7d8e2fb282c714b696f"} +{"id": "train_042436", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e59680cf1f12e89264047abdcb79bc7de45d050d"} +{"id": "train_042437", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ae1cbe2504602d7aaf31c3a4f7967f6aa5b50722"} +{"id": "train_042438", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "15e25afa907245dc100bfa6f77c59668178ee4b4"} +{"id": "train_042439", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2e2ca5a0c415decc21f59d1fc5354e66ee775777"} +{"id": "train_042440", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a4027367be49d2be55a1aa78b5b97d302d7713e5"} +{"id": "train_042441", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "17b8226cd26ae7ea31e37c2adb746d9f0efa7d4b"} +{"id": "train_042442", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "3ff91086e7ffd0d929635e6a53201e2446832cb2"} +{"id": "train_042443", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "60cc1c87b46fabf678e35e33afb66bdf31b34018"} +{"id": "train_042444", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "38ba6f9455b9c3d77f901fca61c2eb39237e7b01"} +{"id": "train_042445", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "dd3a065c0b6effa8b3e91e29d8e7a171a4c0f4b5"} +{"id": "train_042446", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "eb6e7685725710adc5be99e21abb70d33d041272"} +{"id": "train_042447", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4bdfef78a48acbed494bcb74ded716bdf7bdd9cb"} +{"id": "train_042448", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e2429d6a2dbe7b19463c6d14404c0e75a1c08376"} +{"id": "train_042449", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "144379d37767c9b0c8b609782e359489f634b7ce"} +{"id": "train_042450", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2de24f3018c2218734bca261c99a7f5bbc3af0e3"} +{"id": "train_042451", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "19170cb852b511669c8d932d92896aabafc9e5b9"} +{"id": "train_042452", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0d6f9765b53ab90cf281a231abc6122723585834"} +{"id": "train_042453", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d6f046cdcc5cb605bed15d371f6c158b4f6ee22d"} +{"id": "train_042454", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0d7d9386b2175b90174ecfd435b0f5492d0cea1e"} +{"id": "train_042455", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5fb254dc96de9017db59092a2fbff43001baca95"} +{"id": "train_042456", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0949dddb7fb0a8390c519ad02e57f7ee1d57d9b7"} +{"id": "train_042457", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "788d1c98ec6b3b91d0ee22494a9ec2264208a7a7"} +{"id": "train_042458", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f3af13e8013c8e3c2b46cac429e57baec372cbb0"} +{"id": "train_042459", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ed3c0e0e5219424ae82d92756d317fe035e3fcc3"} +{"id": "train_042460", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "5a0e84491caeb34ec5baefdce0d45948fdc43a31"} +{"id": "train_042461", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "9e3fa9a82b953d6ac33ddad957e965b65871c8df"} +{"id": "train_042462", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "4aa7794709546ff4dda528109e94b9249aca49d6"} +{"id": "train_042463", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a37eda737c0ce6a69db0b4cee11dc7a0c5d8fdb0"} +{"id": "train_042464", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0d7601834335080442244e7a2dcb307f4a6591de"} +{"id": "train_042465", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "61f3b4ad86931bbaf95b86cf0630dfbfb9f9e6fd"} +{"id": "train_042466", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a40931b82e283e90bee6a6553441ca75588b6bd4"} +{"id": "train_042467", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "58a7a00d0e71e8a699e79fea564378280ee3db59"} +{"id": "train_042468", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b6b4482644dbbec0272dda8dfa4168245d1c3e17"} +{"id": "train_042469", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "73ec6a51bb8392e69fdc836bb39bce5ef117f408"} +{"id": "train_042470", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "50d0e2a0a4fca380203099e82c04651d203a66ca"} +{"id": "train_042471", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "931f9583cb3b0f979b432a20d40530deea428dfb"} +{"id": "train_042472", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ce789b865f393dae3a63c1bb6af1ea938ea3314d"} +{"id": "train_042473", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1a8edc7d17cf28b88e8d035805ea9b0959484a49"} +{"id": "train_042474", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8961118b61d6781ccde48162790ae04671879ee1"} +{"id": "train_042475", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6494a422452f104f3365c02debc942ada60eb0a7"} +{"id": "train_042476", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f52a71e1c3de1f3e2344929b674181cd2a0e9952"} +{"id": "train_042477", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c5c696b108f53ab792b44dde3d94aa0e77383609"} +{"id": "train_042478", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "89f6ec044b894fac7663f99a2738d3ea0ca1211f"} +{"id": "train_042479", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "41fdd2f570ec1a06817099ad429e6a8d1ee9fa13"} +{"id": "train_042480", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "bad1f5f5a072b0c472aca938e744b579b49bd011"} +{"id": "train_042481", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8fdaf1e1cde9d491f355560c8924203526adcb06"} +{"id": "train_042482", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "88a480b363c62279c91530e52f477c493d9d66b1"} +{"id": "train_042483", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5d3f3e9073fb7e9178ffcb3f3b6f44c5b1b4a5ac"} +{"id": "train_042484", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "57d8e43237e8cc92f1a9b551b475f97beb950670"} +{"id": "train_042485", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "53c3acdaedf49ab1548c7c224975b24b81f98a7f"} +{"id": "train_042486", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "0f99de0fb16a3197cbebf0b287b8efda19adca98"} +{"id": "train_042487", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e4d11c7c312d2cc0d996e8f519defb9fa21dae36"} +{"id": "train_042488", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "b325d27cc9ce8d8c718e8d692bdbc08491275fdb"} +{"id": "train_042489", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e7710bd9c570d533ca134d2266cee2a4d6854871"} +{"id": "train_042490", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "693567cbd616862f8134258e13d0bd9ac40483bd"} +{"id": "train_042491", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0cbd10495ca9a2d6347014ac99654192d0e10f12"} +{"id": "train_042492", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6182ffc87f7a0a3e1dd6421780b623e0d913e995"} +{"id": "train_042493", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1c803d7d28c9da7d874c41a1048be3a2bcc280db"} +{"id": "train_042494", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2d8a0bfea3a3d06b7ddd3267510c62d313862ba0"} +{"id": "train_042495", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "51ae8e3de7e07719f80c238111b74e861b24713d"} +{"id": "train_042496", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d00025d243b9070094fdb47119bcc222ab463318"} +{"id": "train_042497", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "35d8aac1cdb2981c6728a99cec0ad9f5556848df"} +{"id": "train_042498", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3deb6181b533b6eeaa370c9c6c71c8fd068243cf"} +{"id": "train_042499", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "080f48301329aa21f526b499e35eb250359c78f4"} +{"id": "train_042500", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "00c82f38ba18614f2c8754ea3393db323fd7a072"} +{"id": "train_042501", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ff4bab78e80c9a9d9bd830443d2b123386cce057"} +{"id": "train_042502", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "918a662f54cfc8698fc15b6baa0de1d989c74a16"} +{"id": "train_042503", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5f816bd91dd57cd2508ae906a1a2da2901c1b88e"} +{"id": "train_042504", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "638e49f866b4b6c5bb4afa21331eb5fe8be5fd2e"} +{"id": "train_042505", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "88d9d4b992202bb678b207794c29ba65b01fcfdf"} +{"id": "train_042506", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0d8521f8f3dd38f7349b49e22619c67a6a1ebaee"} +{"id": "train_042507", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d71deaaa62d71b4144410ea580e5cd5c1c61b90b"} +{"id": "train_042508", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8275b991dd8973ede12b740507a7195a0c34e55a"} +{"id": "train_042509", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "eb013b484da060ee300cb0c0e42f62f01c290393"} +{"id": "train_042510", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e71fc58ea984297ce47bb6b63ba07a6ddb5013a7"} +{"id": "train_042511", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2b7f62c91f49dc5a5d7b4fc74504b277cddf1a87"} +{"id": "train_042512", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9aa8e92de14b890db38dd8d940f39773b3044a94"} +{"id": "train_042513", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "48b611c44a4de91890e05d8931b5376e3b57b3a2"} +{"id": "train_042514", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "275001b9e5e090da32f3d10d1485daee1f0cd787"} +{"id": "train_042515", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "bf05a51da8e64f411d1aacd8bad14e4467154d6b"} +{"id": "train_042516", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "89dbeda0b44286a48f5d20edb76ea6d33c896229"} +{"id": "train_042517", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e5bcec3986b73e9dbb69e8443fbf4d939f108ef5"} +{"id": "train_042518", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "cb045b3f32eb71efd938b3db3d001ff6067b2b9c"} +{"id": "train_042519", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f66d154183ab58c869afdcf3a66e7edf93ce0f23"} +{"id": "train_042520", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5309d352891e94e21a82f5deb5467cda73f71afc"} +{"id": "train_042521", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a4ef9d4e5d21430eb4fd1999360587f057cc390b"} +{"id": "train_042522", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "769963f865948bedce8ab54d91f592cb71132b77"} +{"id": "train_042523", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "15fcc40fffb56450f794f3dcf43a9ea486cebe8b"} +{"id": "train_042524", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a368f13ae2a8d36d135c6656eb8ae3e3a9ba58a3"} +{"id": "train_042525", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6dc3ba37b5b4f2b0df7c8145fd6a8b2b5e718d57"} +{"id": "train_042526", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "1fd5fa8caa8067c704d6fc9b3c9ba99597ffd728"} +{"id": "train_042527", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "249a528c01a25d7071321c88079e6a3cace428c1"} +{"id": "train_042528", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "edde062fe33371071b8ec5fa871ce2cc6d40ae7a"} +{"id": "train_042529", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3e3a54628dcc9b87647a73bba2062a8f75105dd3"} +{"id": "train_042530", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2e890f4b5dc0d0c5d11052d3b5e61be9fe2a1f5e"} +{"id": "train_042531", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0a5011672ae7f0e16c830cb415022f86df3daf18"} +{"id": "train_042532", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6441bbeeb081a0a904f147e0f35ef2f52a1572d8"} +{"id": "train_042533", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "90bd26758ee1b66e43eaedb5159161a36bfcb635"} +{"id": "train_042534", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "cd4716cff0aef4216334662e003e3a463503671a"} +{"id": "train_042535", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "008de7b5448f6062e3454869a5130a072f764f33"} +{"id": "train_042536", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "428f962dba31e767c01b38837c9f85c08b8f04a0"} +{"id": "train_042537", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4d90c6310d401b5e6a3720c38ae22cda15d4db94"} +{"id": "train_042538", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4eedab26468a00ea7ded21b3514fad8173334067"} +{"id": "train_042539", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ff0e9ea6fd9134a28a971f7d22161f67ae271872"} +{"id": "train_042540", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "3ae2fc8d581c5b7437435e58aced22bed1db562c"} +{"id": "train_042541", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1b5215ee0d32fd44d3124e9dee101b1e315b6d35"} +{"id": "train_042542", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "07f1f2152ff1ec910f56739badfaf1ec141ed302"} +{"id": "train_042543", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "78c45da103f8b4a7c9323abd068564cb207a7f4c"} +{"id": "train_042544", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "25a030436759f9749221c843665e95566b4dd6a1"} +{"id": "train_042545", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9de1b299023d6dc31b6833110db1f851f6d041ac"} +{"id": "train_042546", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4d2e532b4913d9c7d1ed88d421b5bae265c3e6e5"} +{"id": "train_042547", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "16fa9e082773410e14eba211fc23f684d9d55f59"} +{"id": "train_042548", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "bd2e8d705f09e9cd889039586ac6b4a005eba75b"} +{"id": "train_042549", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "bc88129c6909af849770a79492889507c3bde707"} +{"id": "train_042550", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5386b1fab5b5ff39d4fef08a5e4fdf26ef6247ef"} +{"id": "train_042551", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1b7d66d88e4966db25cff24dee632c7156e2e7f8"} +{"id": "train_042552", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f205f4db6a06c571548a2fe5225dd09dac616910"} +{"id": "train_042553", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2076406f7fd2e6f3ca5496920d5121d3e3a9e1e4"} +{"id": "train_042554", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7739b09e725d1ec530e97557e06cfe326531a94c"} +{"id": "train_042555", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c264fa2091438a81262eedd27aea7fcc49b70055"} +{"id": "train_042556", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "485d970d886988a4a926cdf00196d059c4768aad"} +{"id": "train_042557", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6b033b809ad11ea0fd288ccc28ce03d3d255fb19"} +{"id": "train_042558", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2c4f25b84cdbd3b4a8664c455d9d596a8dc0a643"} +{"id": "train_042559", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f1144cd547f6738929fc54da2b1674c7ecccfd5a"} +{"id": "train_042560", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fdbde1e1f8923508b301e49b8cd52894e503aa39"} +{"id": "train_042561", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "53804991b1642514958faed52d7d0429fab1cb4a"} +{"id": "train_042562", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "733d53baba355d96187746fc4901949dce111cdf"} +{"id": "train_042563", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f4da8445aa03e58d11c569b3754eabf174c4a399"} +{"id": "train_042564", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "adc954d11411e97725fdf91972ad670dd5996b54"} +{"id": "train_042565", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "016ee965bf128ab58386bd98e14d6761f58de5e7"} +{"id": "train_042566", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "afbc42b0d5d3099c5ce37275167fba8a1e33bf9c"} +{"id": "train_042567", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2e8d6fbd474a9bc5b29ea24fcbbf7f625608e6ec"} +{"id": "train_042568", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "793e82d2a09770fe2de3ee547fb41981aef0d633"} +{"id": "train_042569", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "47d3da19ce702c5c99552914e2ee51ca620f77a9"} +{"id": "train_042570", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "24c845e2723af903dced58c482deca67b6b02e8c"} +{"id": "train_042571", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "96815d15f523446226abf0f122741586e1f9f0be"} +{"id": "train_042572", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0f0cb0d4ee14968dc53474d9b73ff359a955ca73"} +{"id": "train_042573", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a6588a053bfe7a14b3f359980987809c93cdf5b7"} +{"id": "train_042574", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "831a5a3e46cdfd5a44a7615ff3bb26e4a2c9d9b4"} +{"id": "train_042575", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "775995559d54a596609aa0440fafac171ad4fce5"} +{"id": "train_042576", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1ebdbf509b733aa22349026fdeaba9d939298494"} +{"id": "train_042577", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1878141ab7426c86fc4a629039a33c87119e02a0"} +{"id": "train_042578", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e6e5e775fdb3c0f12eb1507e005db5ffebf5f41d"} +{"id": "train_042579", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "1768f522d457327c9e03c9f23c016234e5ab1640"} +{"id": "train_042580", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "bcac6d0fbee6cf2a0b4fa1b7579788ccd4784bbd"} +{"id": "train_042581", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2c034c64c0c4c206b0481b8a2ac9f0bc593c5ba2"} +{"id": "train_042582", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "23694eb02937c77ec3066a552a4a2976ca0e6f35"} +{"id": "train_042583", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "515d8a54b264eeaa4d8bac52dcf8ec7a9a29c6af"} +{"id": "train_042584", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8ce48b8ec65ae1c25a2f7246564eef11be8a2872"} +{"id": "train_042585", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "dc3a2d58f7fb0baae7063f1f00cef9ebb4d84a40"} +{"id": "train_042586", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7cfe96af3fe0958b400b986730bb5f05294986bb"} +{"id": "train_042587", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "02635bd0f96ddcc1d5896f4f94972bd1586272c5"} +{"id": "train_042588", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "7acbd9292c3663db247ad9e55047bfcd3ecf3fa1"} +{"id": "train_042589", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "bdd0b471fa22e1624ff845186e91f4728e7dd9dd"} +{"id": "train_042590", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b4312190f435ea9caf2fd4ed4c42c4c87e0a31a6"} +{"id": "train_042591", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "485d970d886988a4a926cdf00196d059c4768aad"} +{"id": "train_042592", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "26651d3078515750d2155b3deab076bff65f330f"} +{"id": "train_042593", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7698f78853f6d8c865b87b98c35e520d82983016"} +{"id": "train_042594", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f1902a4fe5738b13fed8b0b86704248af9ce007b"} +{"id": "train_042595", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "67e484ec47b691347d0519615f83306daf855ba3"} +{"id": "train_042596", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d1a11bedfea6827f284bdd72767036746e0200f5"} +{"id": "train_042597", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0a6840d4ddad75e0d364d52255cde21f60850a72"} +{"id": "train_042598", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "fc846ac51d3912c02579fb1d3ad443a9b7dcacc4"} +{"id": "train_042599", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c74f3c2abcc3efb6e290bc961b34307d29b0dcf7"} +{"id": "train_042600", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e3bbede49aa4e87e74a69308440fc2810e69d8f3"} +{"id": "train_042601", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "facd6496418fdda95748fc86e297599618c00e54"} +{"id": "train_042602", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "07f3a9b4d022bbc26b1393641d3fe76cbf884e80"} +{"id": "train_042603", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "eeffedbcd8708460487cffbcab21836bfbb3c2f0"} +{"id": "train_042604", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e0100b1a6c246288a06a170f332ba469d1e985ca"} +{"id": "train_042605", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "dad9765a84dea011fe271d99af64c6b8535a47a9"} +{"id": "train_042606", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3f9a1e209a34b3d273c5397481b9694e6006483a"} +{"id": "train_042607", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3b6e763b446f9b8ec1a9383d212d08e78ca7e4c2"} +{"id": "train_042608", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "bc943e683993d98b4b55781c57c7bd34da0e3000"} +{"id": "train_042609", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "840fc8753486f7854a44c3ceba363d2a39817460"} +{"id": "train_042610", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "899e17fc08fb8c98778b0e7617b6f2ddbb229f8f"} +{"id": "train_042611", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1205ed22b562cfd40b52acfa7d23f9f63afac682"} +{"id": "train_042612", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3959520789d8dee89caee9c042938dfc015f96fe"} +{"id": "train_042613", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5210f23153360b52e84176590f9652d9bfd8efdb"} +{"id": "train_042614", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d0fac72958b9437ad3c87f8cf75a4e66044c4eea"} +{"id": "train_042615", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1a0562f47689df904671c4dcb3d8e06167a82be6"} +{"id": "train_042616", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "90bd26758ee1b66e43eaedb5159161a36bfcb635"} +{"id": "train_042617", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "974317bf967f6e246a7e80eb6a4d07b2d9226bc9"} +{"id": "train_042618", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8f8cf64034a461cc48adc43acf20f21eeffece05"} +{"id": "train_042619", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7be4b5a9c5d2838276b22fe7a789c30621c07b6e"} +{"id": "train_042620", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "61d5e50e320a5131a20292355cecbf1d1524f3ed"} +{"id": "train_042621", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8269f02a66cddbe595d1f348a86ba1d171a97460"} +{"id": "train_042622", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8d45ee23d46bcbca05a155d295992ddb3722b1db"} +{"id": "train_042623", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9b104a63e37b326c75315c911c319d57c53ba905"} +{"id": "train_042624", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1c20916063afd738e4a11e7e970dfc41e47a40ab"} +{"id": "train_042625", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0e609322e07e53098effa126d38934300b0a06e8"} +{"id": "train_042626", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "aeaf616d20c95e1aea1ac2a376289071e2ba4eb0"} +{"id": "train_042627", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e87c9aaba1e91ce3f639f1ee38f9dca1a285cb92"} +{"id": "train_042628", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7130dc4d7d51c9bad96c4b15a84c19a2f0337eb6"} +{"id": "train_042629", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7b63c669cae428cb8c9b9f1f8d8563680cf14ac2"} +{"id": "train_042630", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "45ad3f39291c94740226cb1d1eb8e33253d7abc6"} +{"id": "train_042631", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "83a8b89c3f8b337fe1bedbb3faf273c336ee4015"} +{"id": "train_042632", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c0a02d75a52de834bfcbfd9734d3900d64e8e234"} +{"id": "train_042633", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "25a9759a861baf7cf3249643b965ccf812d3aa2e"} +{"id": "train_042634", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "67feef66bf10d67d2e3595d255812cec32069f6a"} +{"id": "train_042635", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e59680cf1f12e89264047abdcb79bc7de45d050d"} +{"id": "train_042636", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "136bb8fcc51a9a6e3f38ef0f7950fc31e518de55"} +{"id": "train_042637", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fcde83f5e8f887ad441d89e7ff444d0a2d586963"} +{"id": "train_042638", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "174120f7c063836b65ccf5545521a7d5c924bc3d"} +{"id": "train_042639", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6fde991fd3d4c5e1917c0bc2f93527301c0572d2"} +{"id": "train_042640", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ee933905944756175711744a5552640621c7d002"} +{"id": "train_042641", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "09ce332e093c3a895b33c24b7ba6be96685fad4e"} +{"id": "train_042642", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d445a89f648bb01a4153450bffab195623ca7b8e"} +{"id": "train_042643", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "194d2e90a513ce082cde373d29f8057b0ef0cf04"} +{"id": "train_042644", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "225fae2d776e01af48350decfac733bcb97a8ebb"} +{"id": "train_042645", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3fbd33441a8945d51a22a97e3ddbe368a3c28994"} +{"id": "train_042646", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "046e73f9591a3d975af71d9c856f8dead69b7c89"} +{"id": "train_042647", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "34bb83c7e6f2c40f5d637ed93c2f9f0e38cd5151"} +{"id": "train_042648", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "26ccc06e63a077472482b0d39e2ee4289eedb19b"} +{"id": "train_042649", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "10efeb6562ac55455023e67150b49013198f5636"} +{"id": "train_042650", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5ee2f63f476091c3698d4882d180f25cbe9f6b15"} +{"id": "train_042651", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "32b5704c93a7bb9297d13d6804c3d5729fc52e0d"} +{"id": "train_042652", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "95692c01b90bdbf0daa65cedd49ad8abde0491f7"} +{"id": "train_042653", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "924a9e2a08a65727c35776dbc0508607fcac9a66"} +{"id": "train_042654", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "545706c78a87636c6f6857b37112590082ee1d76"} +{"id": "train_042655", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3d2881a15b669737cb34953f1f4e276ea1eeaf85"} +{"id": "train_042656", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "181632fe9ed49767c8d5486a632fcf6549bcc785"} +{"id": "train_042657", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c1f42b56675e7a3fcf2de25924652fd6fe0ebd6f"} +{"id": "train_042658", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1275c49c89da9aa3be3c98ad53717940d5622fe2"} +{"id": "train_042659", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e9f395021e43a7542908a4e60379d7e31b5410d0"} +{"id": "train_042660", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9b20e869fca595cf7d7b48f5a3145db6590fc423"} +{"id": "train_042661", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9af9562d7879b67adf2f1126ee1c7cb3e7de2c16"} +{"id": "train_042662", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "61f9ab06efa717989443901437f06ad94750a352"} +{"id": "train_042663", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e66c8cb80a71c057b7a0862d4151d9a1f1cc8585"} +{"id": "train_042664", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a65c52c56a3b00da5623ad2dcdb344c141c69fc7"} +{"id": "train_042665", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "db9329294ac1d5e2be9ef213d00176830f8c6ea3"} +{"id": "train_042666", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f18fad4f73ccbcda4b2764cf44a144a829595a8f"} +{"id": "train_042667", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "74b7766143b188da1b2053f36f37f514681c3daf"} +{"id": "train_042668", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8bb2fea9655fb2ea835b190c026540cda7bd61ef"} +{"id": "train_042669", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fe408ab6b3d1a97730dd41cb26aaf4848f47c505"} +{"id": "train_042670", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2beb0fd7f236bf51d628a6d2d09707598ccd9d8f"} +{"id": "train_042671", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "680ba1b60a2bc8a764056273849b0e7f94c40db2"} +{"id": "train_042672", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ece02fea245f32e47ce3a4019a572ae9f82ccbd9"} +{"id": "train_042673", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e87cc6c2f152e785dbb38c6966b1c420765b2461"} +{"id": "train_042674", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "25baceac7043f47728e733e4e8a6d1e17876d7e1"} +{"id": "train_042675", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "09f99163b03ed78db592008b9c57d789cd519fc2"} +{"id": "train_042676", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "aaca6ef8735993247caf475c8df5fcb1fe434d9d"} +{"id": "train_042677", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "29cbd4deae69394b34eea2d25c010256aad71d75"} +{"id": "train_042678", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "01b888c77840d2684943835b83fee7a23b64e88b"} +{"id": "train_042679", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e7324c302dc6640a87b0fbcc99fb3bd5e99f9970"} +{"id": "train_042680", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b94d98d1722d18c5f302b6435ddf023d961b3b58"} +{"id": "train_042681", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b0b6a3b7d4ca88a23616b1a4ab385f7dc2fba460"} +{"id": "train_042682", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0413f19ee9e35265b697c7d2e6e622a5623db24e"} +{"id": "train_042683", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "af5a4d1dd3cffb61f026597dd91024b7f40f9395"} +{"id": "train_042684", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "df2a9d4a415d3e022001b9ec1241b11e55d0f015"} +{"id": "train_042685", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ba2f219adf5acd892502349212acf7c3c1561c47"} +{"id": "train_042686", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "de363579a6f4c64e956e6baaafeaf8acc095ba8f"} +{"id": "train_042687", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e1ea0a57e58c19ac65cb363150e5b4fcedc06906"} +{"id": "train_042688", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e31758e432ae01de2139845375b6ec8b47a1849c"} +{"id": "train_042689", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "06d227914ecdce2a59159ab0cd474f49d3e8bc23"} +{"id": "train_042690", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2ffdf2d1c11df56ecc6a51d009a131c24da9153e"} +{"id": "train_042691", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cbbc23f1d89d0f0387729a4f6ea3031714dc6f40"} +{"id": "train_042692", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a70dc29aec3d6c9cde658cc768ef7c574aa979fe"} +{"id": "train_042693", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fc6b9dd11fa7ccf7ced6b508984dd37dadf0e604"} +{"id": "train_042694", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e867d6213089449773524c09e8443d68c8448e14"} +{"id": "train_042695", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0fd6706d5940ddd76d57cde1d5976d3e7608bd0c"} +{"id": "train_042696", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d7a0ca12ec36583f9513cf6b331c2abe18e10ef3"} +{"id": "train_042697", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "57a828923fae4c41de42fcbda96e0a154361cca6"} +{"id": "train_042698", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "104a5b3a6d07fa556f71d70476d93b317ad83184"} +{"id": "train_042699", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "3d8a5bf1db7689c74f8a2408eefaf6238d93c1f3"} +{"id": "train_042700", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6d8b938f1a7455947d8aaf2348c7b70354eb06f8"} +{"id": "train_042701", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1230e76b7f4704965b3e9dd6d666990d0a81a8f0"} +{"id": "train_042702", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c4c67e811942317b762b995d80e790ad55711ebf"} +{"id": "train_042703", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3f45624d9e786819793f69ec97e6eae807adbe6b"} +{"id": "train_042704", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "361fe4ab921dceeb91530c44a2be934154114336"} +{"id": "train_042705", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0b823a34f41743ef0a494798650135c5693ca92c"} +{"id": "train_042706", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7d8c952ee8239e846c43fc3751c4bce60a092580"} +{"id": "train_042707", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d90cfc8842f397998353974b2f770ece667b8549"} +{"id": "train_042708", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e0f8d254e8671da347d69a87208532ce9d91236f"} +{"id": "train_042709", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "013d49b9bc8f67c3ffda6ad23c50dfbd0a9273d1"} +{"id": "train_042710", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e52242f253bd33d04b2f93bc16bef2ae0146f01f"} +{"id": "train_042711", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9970e93531d0f8a8504319d3e4a63b4d30d320fa"} +{"id": "train_042712", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "92ec9e0323b13ab4bc5dd9a629a389122327250d"} +{"id": "train_042713", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b85f67221aaee8f5e4fa9be853851e2dfcb00632"} +{"id": "train_042714", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "cd92fb80ee3f1abb1fb8fccf6f89f985fb3af129"} +{"id": "train_042715", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "678ae6324e32b2ea92e1005605a87384f785fa4e"} +{"id": "train_042716", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f832b375c4e6287443581f686e46f7f9cc556f21"} +{"id": "train_042717", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "1f2a8b5bdfdbeb726965a77b22f3311152dad1d6"} +{"id": "train_042718", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5c74e3e3abf721fd7308e9f84e4359f937e08ae9"} +{"id": "train_042719", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4bf336d50b5b63bd245f47a4626595f3f63a1ecc"} +{"id": "train_042720", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b4f98b0c54a6b3935d3ddadd05666a4a42fb6ac2"} +{"id": "train_042721", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "889b9753d9b25c6ac9c6930fb104713d5715bf2a"} +{"id": "train_042722", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d6d6c112f14e394f6e79a6d51c6a67b9b77624b7"} +{"id": "train_042723", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "b6ae8f6d16af2a44cdda40ee8f12d26c6350ba6f"} +{"id": "train_042724", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6b6161f68a89e9018bd3b63951c7dad47d59b0ab"} +{"id": "train_042725", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4dbdc64fab270ac9e1de9fda6c6f019bee295462"} +{"id": "train_042726", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "16cf1267b788fba85505b751de14603523400139"} +{"id": "train_042727", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "c664fd32416037e33a135585cb20ad57b525ccae"} +{"id": "train_042728", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "424e37c8a4c0ef131d37e6f5c3b548cef8b087fd"} +{"id": "train_042729", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2517075c8d30d52b71d86b53e04e79e1a97d2e1a"} +{"id": "train_042730", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a78f456a8c0a8140838528e7d02f300e9d7e6d04"} +{"id": "train_042731", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "bd52cca696f51c96a25df6a13c2f4a79672a9f9c"} +{"id": "train_042732", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "93dfb48e5134a8c8c02a74f22401e9951a900737"} +{"id": "train_042733", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "15fbd6070fce07a07480c2b55fa0ceb0c3d79b11"} +{"id": "train_042734", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d31df216313737f70b544c5c6cdbb5eee38a3ad9"} +{"id": "train_042735", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8116c679af5fd0d79bb2158b47fad006ad826e0f"} +{"id": "train_042736", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c3e01fbe2ed71ef68163ad090cc72902a115f5ae"} +{"id": "train_042737", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ac9fc62e99a5136b59690f7a28c165eb9ceb7942"} +{"id": "train_042738", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ee940308d8d8643a517d6fb4bf9fa30c65d2a276"} +{"id": "train_042739", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "bc319dcc138471c8dba646c25d7fce7c1f01218c"} +{"id": "train_042740", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4eb1063cd49d27ddcd8620aa8cc19eba3c560a46"} +{"id": "train_042741", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "123798e5b579189329316b8067ae73adaa26287b"} +{"id": "train_042742", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4ea87e7d9dec5232ebf98a69bd19315bbf73d45b"} +{"id": "train_042743", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "945e8dc7e7b6e4b2bc772de6c2b3d2c21939f136"} +{"id": "train_042744", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "67301215ccecd0b60cf19135ad9b37dc801a1091"} +{"id": "train_042745", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "1850c90da7c75f5d0958bfecadde00e7474e52e0"} +{"id": "train_042746", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "dce93b042cfa2518401973623e98b1080781fc8b"} +{"id": "train_042747", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c24be61878892ba361e5566a71fcf6d4612f3723"} +{"id": "train_042748", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7da8ca702e988d7ba4ee703b06f4256b367f728b"} +{"id": "train_042749", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "59d46c50126b82f2e56207b55a72cb7e535dc36e"} +{"id": "train_042750", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "160463cee8d58546990e1076072b806b3d14e76b"} +{"id": "train_042751", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "865563eff409c09392fda94e5cbe62c31334a310"} +{"id": "train_042752", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "41791728e1db3b6ab26e58d39eff9c9b41d8a0c9"} +{"id": "train_042753", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c5c696b108f53ab792b44dde3d94aa0e77383609"} +{"id": "train_042754", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "80aec6b05e14c68904e3159f775e800319377166"} +{"id": "train_042755", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "db2bfc36a31f333a50dbed2d990f76cb0c14af6d"} +{"id": "train_042756", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ee86bcceb079990cc9efcea024c73c2758a70139"} +{"id": "train_042757", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1e8a177a033ae5877fa801c8bcec762ca3da3bcb"} +{"id": "train_042758", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0c41ce97c9eea4d09dd4f103f820d1b8b8901391"} +{"id": "train_042759", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "18e9d61665249cbe98df7a8dd4eb34c5c94617ec"} +{"id": "train_042760", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9c18e4083f1e5e9ed8f49bab8f7f6335227fcbab"} +{"id": "train_042761", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0bd48fcfc05fc7a17cfc142151e9f97584293645"} +{"id": "train_042762", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "7fa3a8fa476aa00b0fbbc91b26a9d76eccda9a63"} +{"id": "train_042763", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6ed063afb80d0da179d974c628701f67b2f91836"} +{"id": "train_042764", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "73b7a2d948067281e77bf0e75d36f49594c7d245"} +{"id": "train_042765", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7b74d6320a1c9591dc92fa09dc0d6daaa580967b"} +{"id": "train_042766", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f96fe65589cd7c473c4282d77572551ebf9ece6f"} +{"id": "train_042767", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1eedfadb1b3afc2d49136574bd522c5a77e9dc6e"} +{"id": "train_042768", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d5e4c879818a76b1606b14f341f6effed790fc68"} +{"id": "train_042769", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5063a9c81390005ee2d4a0ff40f1352b7be53ac2"} +{"id": "train_042770", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "82848572e0099984e3e26c77f26e773ceb588d40"} +{"id": "train_042771", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2104eeee33598c56abc3b3b2284e8298e90e8a3c"} +{"id": "train_042772", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6710e9f7ba3d23c5eb0f40c992ac9c59737914fe"} +{"id": "train_042773", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1faf6254e00a4eca538122ecbf563bac31b4d606"} +{"id": "train_042774", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "35e6d34fe8e2444863946517ff7d445f190b29c2"} +{"id": "train_042775", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "05d0d1d9dfa8d9d427de25836719196572c1d1c0"} +{"id": "train_042776", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "06680b39145e083d9544afe1693951947d33f42e"} +{"id": "train_042777", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "00176edef004ffa2eb91253efd9adde9fa7893f2"} +{"id": "train_042778", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3ca3f3909b24989958bf1f3b3550b87f2e762ecf"} +{"id": "train_042779", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a56502e7f85e298afbb05c15d70cf08bc1aff974"} +{"id": "train_042780", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "0d1be43138809b7a0b93b1ef908894a81788c356"} +{"id": "train_042781", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "82f60e379632301880ed576c2d03943fab92399e"} +{"id": "train_042782", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9b41bdd138842e530d443fd6c8b44fa73c526541"} +{"id": "train_042783", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "04e20fbc1ea1a0a66e3dd3fa17e51172430d7e47"} +{"id": "train_042784", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ad0fbbbc69f3a7bfc32ec147a1bebb98c2343a1a"} +{"id": "train_042785", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7cab9722958ccd0f4d6c8b1e126ce57ece0471bf"} +{"id": "train_042786", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "23216ee10e5b2cdbaf4d4a27ec357e53ba01a9b6"} +{"id": "train_042787", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d840fd421d029ea2ceefd288f020280dcd6f2b1e"} +{"id": "train_042788", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "18cdcd4b48fd64129e42d1d18edf2d8a985ec520"} +{"id": "train_042789", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9f6949dbe90ea9a1a37ac81e78adc70256ebcfc6"} +{"id": "train_042790", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f87232cd658c0a3cfd8af758248fa96cd5e5cfce"} +{"id": "train_042791", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "04e1bd16e41cea452134a092d1e8d684227b79b5"} +{"id": "train_042792", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ffa5e5eb3c22df65fbc7246e8ff0127530e5b3e0"} +{"id": "train_042793", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "54e30f6e0376d92c821dc04d9b56cee37b3308bf"} +{"id": "train_042794", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a776c6c813064fdb075bcc9f209e13b871cd795b"} +{"id": "train_042795", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8290af07c4dab78e1eed6406a3831e8f48859118"} +{"id": "train_042796", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6891a0add8c99bc5c02576b71e37990ac423b36e"} +{"id": "train_042797", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e7c61d3fa2e9a5c0e3ad11f8529eee91b80ca719"} +{"id": "train_042798", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4898cf3e38bcec89b1bb91297c92dee10dc46651"} +{"id": "train_042799", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0fb9db30556fa12f004d8ffc7e3596f87aef3fdc"} +{"id": "train_042800", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "c526f9f97ad6ed6d979bcbdb977a3299c7968ae4"} +{"id": "train_042801", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "03cd4a256a075b71a6bc0a41cc578de672f7b815"} +{"id": "train_042802", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "83f60bc4e8f3b1ebe94d17f7e0c22a91089673cc"} +{"id": "train_042803", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "db155495dce08f127ccb3b31c36f36a06809b821"} +{"id": "train_042804", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "632721eede9bb15b14f5e605a1d521bf06c08ce5"} +{"id": "train_042805", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "53804991b1642514958faed52d7d0429fab1cb4a"} +{"id": "train_042806", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8b27cb35dfc4cf7b6005926b6c15db684e89609b"} +{"id": "train_042807", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a7a8c3797a8aaa734d0d13ada49fab53c1d044fd"} +{"id": "train_042808", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "90daafdc1bca93b5184fd6ac210abcf3cb91aad5"} +{"id": "train_042809", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a9954f0d75dd8737c7afca4a030c8500ac04df8d"} +{"id": "train_042810", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c72f18c5eea17a9327ad345e58c9e998f5dbe84e"} +{"id": "train_042811", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4dab14f0ca0e18ba995cb1bdbef2015a7f0f2ca1"} +{"id": "train_042812", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "93b341d3ea3406b577868f7f474299562e9dbfa6"} +{"id": "train_042813", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "df6ef9e4121c335766af941ddd7528b3e4be4083"} +{"id": "train_042814", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b072b044ca78c17a848e71fab88e20ca59a6770a"} +{"id": "train_042815", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0f76576419b75bfa7dc802b2f213b3e1b541d677"} +{"id": "train_042816", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "cdca2f1b9c93b8c81022fdcc4dca4fac44f09388"} +{"id": "train_042817", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0db5d67d2880674c650df7a31e5bf14785b06f2f"} +{"id": "train_042818", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "82a65f2d5f9dc7743f97505c6a72a707eae90f62"} +{"id": "train_042819", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "45c4b36bb3d5f702e102108b7bf299f89e51b19e"} +{"id": "train_042820", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "eb36ba23ffce6b72378e07cc69df7b91e45dd44b"} +{"id": "train_042821", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "4222d8f35d53af0df527229f9c3677f705a91f68"} +{"id": "train_042822", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "cbcec364ed384ced79028c2239e7e2a957d020e3"} +{"id": "train_042823", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "58248f30d04c57ce1f9d1a5397d17ba7fb86f5d2"} +{"id": "train_042824", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f4ee5b4debc8bd5d6ba94ffd7c5c6366968fd5af"} +{"id": "train_042825", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cdc6ae5472be1ef424c49771a77ace99c76aa89c"} +{"id": "train_042826", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7844960ed0371268fe5e4b86e6194a60bedc15a0"} +{"id": "train_042827", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3137be50a0ca59c148a21415872caef86a766e03"} +{"id": "train_042828", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d5fb5c9d208f692cfa60d25feb941ed1eb81a7c7"} +{"id": "train_042829", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e7a83e79356d457658dec0f88e64cb03e16097e7"} +{"id": "train_042830", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8cc9f845bb4d060d5b1f1b331ac357c107caddeb"} +{"id": "train_042831", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "12d304eef62004a677ab49dd07a0dbd936f0ff71"} +{"id": "train_042832", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "73954516cd7708a9fa02050a711814fce64daf12"} +{"id": "train_042833", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e9e477faf1712149c7e8b8ee01018130db0e74cf"} +{"id": "train_042834", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ccd1d9da952c4552e6d1d84804a7a3d6dc133007"} +{"id": "train_042835", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6c8f6db6f5af7bfe0ebf9e5fe36df874e7477944"} +{"id": "train_042836", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "444c87e3f0fe13a5a3d34266be8d7f4dc8c738a4"} +{"id": "train_042837", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9083bde29cb2be7868c4be850d864904b3f42df6"} +{"id": "train_042838", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d68174ba28b38affa5838686d59a8189051b7f4c"} +{"id": "train_042839", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "cb2136ccdce9db0aec891e05aca32004627c3c94"} +{"id": "train_042840", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ca9e2c24fdaf10cef8e76210c497fab7f05a6c8e"} +{"id": "train_042841", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "cec172dfa3dc12f0136923e5bac01b15030830af"} +{"id": "train_042842", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "13c35c91526621cc0b97c4a8c79ab7a81b4592cb"} +{"id": "train_042843", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d389cf0ca8d7b15c05f2d5ef01adc9e3e15958cb"} +{"id": "train_042844", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9a6994c144e7077a25d1548d8964bef2c05386d9"} +{"id": "train_042845", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "1230e76b7f4704965b3e9dd6d666990d0a81a8f0"} +{"id": "train_042846", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f19b3b3c5d68087f81879a81ec706dd0d7d14f63"} +{"id": "train_042847", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "49ad207012c3f6a922a903708b064e69abe32840"} +{"id": "train_042848", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d6ef3a50c67895e0e191b72bd840f2255c8a9cbb"} +{"id": "train_042849", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "57e1a2f18f725d2cba3d59492884cfdc6ec7920f"} +{"id": "train_042850", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a3568cfbb6ababd1f36b53a035d854778d273675"} +{"id": "train_042851", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "11c1d09e71b4877f31d4c13e8a4521886d95c34b"} +{"id": "train_042852", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f1d8a04b4541d9f5760a26306e4c3dd2627a1b06"} +{"id": "train_042853", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "3df41aac8f6a0049d88f195b41e31c463b3cd8e7"} +{"id": "train_042854", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2f5382c1a891bb43f4a202c1f0472540c2f863fa"} +{"id": "train_042855", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e8c59c7e5e824de0d9ea6255f971776f9e8c3c0c"} +{"id": "train_042856", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "97ac702dd2621cad7d3080c83406e91ee0ee6500"} +{"id": "train_042857", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9ca262315964e1bb5d02b21836c5054a335bce58"} +{"id": "train_042858", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "626fdab58f2e27525f312243ff24f91c2eebf3bc"} +{"id": "train_042859", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "d17a1af3ce29ffacee977c920568b9b00940b07a"} +{"id": "train_042860", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "cfbca6aeaf20f5cbfa819f8e701718b53b523830"} +{"id": "train_042861", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8f261c80109e98f0a89799f189fea8ab0581a22d"} +{"id": "train_042862", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "23526f0d25af6691ac46a1088b2d3e3f721388c8"} +{"id": "train_042863", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "407f7109055df6cc11c5d108575710f5d3b5b9d7"} +{"id": "train_042864", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4e21352a2615381dd4bc7c7457888d20aa37fa6e"} +{"id": "train_042865", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "df5abc138d8b72b39ab2b807c3c56c25a7b57ec2"} +{"id": "train_042866", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "51aaad3eac4d8cc62fb0873dafb426deff885b90"} +{"id": "train_042867", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c8c2175ae40cd4bbb10e30c6ce26417a43a3e297"} +{"id": "train_042868", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "edebe45e98f68dbae8031c13e94521a3ff0adb99"} +{"id": "train_042869", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "55a78e25197544b6df4e2beed67c5d146ed3f906"} +{"id": "train_042870", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b1d058e4522d9a3b5eeb051ad88df235e6f1321e"} +{"id": "train_042871", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c3eebf54a00725e8cfc14e0cfa21d42c0ae5347c"} +{"id": "train_042872", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "dc8e18987218e59059c31552d39394ab95e65067"} +{"id": "train_042873", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5cc86bb599de4afe9cf6bd5599fb27766bda4be4"} +{"id": "train_042874", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "89bba04ec4415a551e1266e29944783cfdfc884b"} +{"id": "train_042875", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "102bb287390a76f77c565bc07aab9d58e77f3557"} +{"id": "train_042876", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "98da5adee67d32c57c66a53afa43ed3f6be67f56"} +{"id": "train_042877", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d53ee6e5fff4ed52804da1539e1d83d503ef9d6d"} +{"id": "train_042878", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2ba3baa114f1788261f950dca4c271ab47d37a12"} +{"id": "train_042879", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "730a9a055557999f12d38621d506dd69dc95f7c8"} +{"id": "train_042880", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "95a9c0993f90565b67341948a14c536c4dacb0c9"} +{"id": "train_042881", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0e096a8c8df7991b6325f9e84e25c037f5a274d1"} +{"id": "train_042882", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "b1443d496473e621b0be8bf49d7e8403a9a25041"} +{"id": "train_042883", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2ae9ef6eff7419ed9dabbc91bb78047b0966effc"} +{"id": "train_042884", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0d0fc098e6fb72a1c3441fa94421c4ef168a968b"} +{"id": "train_042885", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "83b4d918c838f2ee8ee5cad82795f5613781a5b4"} +{"id": "train_042886", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "818c8576820706624e82ebf30a6b5ca4632d65d3"} +{"id": "train_042887", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2631ecc998392e08f989f5d890a206ebda750e13"} +{"id": "train_042888", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "20a0f4600d31969dcdfc1a09b81a31a8c09f8c1a"} +{"id": "train_042889", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a730905bc81e54ece1c8902705f5e1d9fbd21f4b"} +{"id": "train_042890", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8d3bcb8cc9cd92ad9f3761ae992dbd38f335af99"} +{"id": "train_042891", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1ac95a99261302015ece116b933e4ea70d53303e"} +{"id": "train_042892", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9e66d32f015ad60500f5063f03536e006b513a30"} +{"id": "train_042893", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "20d59dd3721ae817e73c704c83341bcd7d40c942"} +{"id": "train_042894", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "71e8a89e355d431e904ecd4e271c7ec13f938296"} +{"id": "train_042895", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "44ec0b81dc33dd1654005d0997578d07164e9c46"} +{"id": "train_042896", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "da187dc434fad34d99ea81122701cbaeabb2e175"} +{"id": "train_042897", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e8a421c107679864f613d238614f4f9b5d907266"} +{"id": "train_042898", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "df52cfb54d55529430fb514ba61d5c815271b746"} +{"id": "train_042899", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "3c0b07085272557e1446baba0c4c52874722457a"} +{"id": "train_042900", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a377deb240bca26b7e614468897c305c85a8084d"} +{"id": "train_042901", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "38dcf1bd1d286859a41f9455b4f729d532339560"} +{"id": "train_042902", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e899b8d13ba76bfc33aa06fb858e1309cfd8b804"} +{"id": "train_042903", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8cd67e0d37ea24a8575ab3b826265d70855c9253"} +{"id": "train_042904", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "679555ce63be3a66be792c03d93ecd36d5c924e1"} +{"id": "train_042905", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "27a894500eca3ca7632dc1a7f116fcdc102f21a9"} +{"id": "train_042906", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d1ed87dc1fe5687f378932a827b4aca26386c152"} +{"id": "train_042907", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "7c4fec3d89ee584fbf021d6cc9ece1c92bdfda65"} +{"id": "train_042908", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f5b86838b5f8eb264b88fb0728fdbdd925878ac8"} +{"id": "train_042909", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "afec8172a1669851b8742a2a612491da6662bd5b"} +{"id": "train_042910", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b07c397dac5f77a711e915c58430dc7d02cc42f4"} +{"id": "train_042911", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "9fe9641397be92ef9f562b2b9a2a454b48d66c67"} +{"id": "train_042912", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f76bc4a772d07d21b5e821457936f0b12a91df29"} +{"id": "train_042913", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8b34b2477481106da00d7cb55dcacaa7a0502b08"} +{"id": "train_042914", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ac4a67329e64ec29087ae7d89ac7d138339895c9"} +{"id": "train_042915", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "608de79249dcb385ccfe203ee4bd2197be646b4d"} +{"id": "train_042916", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c6e62217185c011d5668efd84767d17412869528"} +{"id": "train_042917", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e16ab66c8f0ff3d4f46b8389132faa50bcf0f87a"} +{"id": "train_042918", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "618ad0ee1875d8df538bed5c8e474f8ad874a053"} +{"id": "train_042919", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f1b7fa26cd96c699a8ccaf19a145b20e266a1871"} +{"id": "train_042920", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "31ebd35fbba4b36cb9f37f911ee7d9904c49fe96"} +{"id": "train_042921", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f7432f2f66f979462af605c787636ac3ee09fa79"} +{"id": "train_042922", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "818c46d8a98b8d6f7e16aac6e7a608b1431f20a1"} +{"id": "train_042923", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c42b10818464b0bb07158f14e167b002b5e29c64"} +{"id": "train_042924", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "67845c4bc98ba0a5a0295a06198ad6c49eef575e"} +{"id": "train_042925", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7d9c30cac8f4adeda347ce5c63ec816290068438"} +{"id": "train_042926", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "319a6ae90ca1ebcc56a7a7b715bc2c90bfac7f67"} +{"id": "train_042927", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "25bb33755b010cb03f07fcd06230a37ebb60fb22"} +{"id": "train_042928", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "11daa22d8d82eb9df3fe1700ef19b9a69f993052"} +{"id": "train_042929", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c962a028f02a598da79ffa3dcc3178b446d0133a"} +{"id": "train_042930", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "42dd6a8d5d9c2c50380e1cf555c9d2e2c72139c5"} +{"id": "train_042931", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "28adc77284b821e36ef77c49076e7bda60e7d42f"} +{"id": "train_042932", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "665ed4b4e0b8742b231de0407160a2adbf2d4600"} +{"id": "train_042933", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "539b7d8f709f31ce8cd27ddb9cf102f8ad7fb5d2"} +{"id": "train_042934", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "43f19e773e9e5f1ec8b8c4ef47e6d6b1343946ac"} +{"id": "train_042935", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d11fcce096689e99aabcaaa7157cecce7ab9e842"} +{"id": "train_042936", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d56920b4b6b22cbc4aba993d249e883f242ebb61"} +{"id": "train_042937", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "7f7194a6f43fb36c7883cf133fac4168b8658fb9"} +{"id": "train_042938", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8cc9f845bb4d060d5b1f1b331ac357c107caddeb"} +{"id": "train_042939", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6b9c6ddb79098b381cd8b50e00c98b2005962e42"} +{"id": "train_042940", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "7b72e195903a705eace20b82ef4748a712da0fd0"} +{"id": "train_042941", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1ea828bb4d69db9d61e8b3b2bec957b8752b32b1"} +{"id": "train_042942", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "65620079a7a1dbce8f16975e68b6c3356dd4f672"} +{"id": "train_042943", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1b872006ec9ed8e4deae0a4b16555422e0659402"} +{"id": "train_042944", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "711f4799ff1519d217c9be0062ef1124ac6acc0f"} +{"id": "train_042945", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "09f99c86c89a565abe648febd2823fab3dcd7786"} +{"id": "train_042946", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c9c58773a3d8c8b7ec9a52db05e33f1a84bd0813"} +{"id": "train_042947", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "05d0d1d9dfa8d9d427de25836719196572c1d1c0"} +{"id": "train_042948", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "916c3bab5d45d0e539e957cee02cf667f92c3d28"} +{"id": "train_042949", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "abb2b5742755f0dd5261ac8ff1bd9ec4b032c0e3"} +{"id": "train_042950", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a0770877c2334a374e08c7830627700460e0d25d"} +{"id": "train_042951", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "21d2e64c08f7781d3b35db7d5de83a1f3c26060d"} +{"id": "train_042952", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "39ed88344489a4e822ac85da450f46bac25e44f2"} +{"id": "train_042953", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2e04bc73b9fd8ba53d63b18eca84c96c6b624606"} +{"id": "train_042954", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5a4b3178382d07b16b5af6aa1633a4f44d56129f"} +{"id": "train_042955", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3f285b94904cd2fe846e9e2a07532a9bf97d0a2d"} +{"id": "train_042956", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "82e7147c9543b21f34d5b73829c823720be91a9f"} +{"id": "train_042957", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "1359545ff3ee2c4539dc0aa09c0290e8e47bade1"} +{"id": "train_042958", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c238b7004ef99bbeec1deca6a788bbd8a541d8c3"} +{"id": "train_042959", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "9ee28e7eee02200798c44d5b315d025bdbcc43e0"} +{"id": "train_042960", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7ebd4a59790ae9b4b6e69feea8842720c81242da"} +{"id": "train_042961", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a40931b82e283e90bee6a6553441ca75588b6bd4"} +{"id": "train_042962", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dc25fe93a776b99d3ef69be28e6473e81f8bda14"} +{"id": "train_042963", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "02d37a8af2c78a9e55ae91974f55cead7942d54a"} +{"id": "train_042964", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "342aa9450a640d8a9d813ac0a65bfcdf33e4db1a"} +{"id": "train_042965", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "7afec8137c4cb5d2cf6592622ba5f9db8ee2e32f"} +{"id": "train_042966", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ecfc0ea36965999ad6ef502f3b900a0397cab290"} +{"id": "train_042967", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a2fe3cfea76edeaded90edb9768366d9e3783a5c"} +{"id": "train_042968", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c264fa2091438a81262eedd27aea7fcc49b70055"} +{"id": "train_042969", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9e3247a9537744deeddc32b8898c142cf6042a87"} +{"id": "train_042970", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a321ff685d63fbaf813550c6fe0057d63b3dccc5"} +{"id": "train_042971", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "486c4853c8bb7ac48756df7375dc6a57a75b5f5b"} +{"id": "train_042972", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4d2e532b4913d9c7d1ed88d421b5bae265c3e6e5"} +{"id": "train_042973", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c5d75035739fc009f291f4ecfa77984185243783"} +{"id": "train_042974", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "228b3e725a0bb2363be9ecc51a5a5469081ef164"} +{"id": "train_042975", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0571f2d09047e0cb113dc127b14dcdbfebe5827a"} +{"id": "train_042976", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cfdca7ef97f9448d91d981ce321c714feff74cc5"} +{"id": "train_042977", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "964df34c39e35360d15ecb13d247d417049d6ab3"} +{"id": "train_042978", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8a1aea630ef7add9ca297e4fb00dd6320cd702da"} +{"id": "train_042979", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "3d50d0388a748d1555bacc72b22cd1dc21988858"} +{"id": "train_042980", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "0ec0f6b82af6153cc096e979b1add1f95aeaa191"} +{"id": "train_042981", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "dda6e81e669c26180c85b0f2204fc2362e87b938"} +{"id": "train_042982", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2e34af567ab130e8fdc81134c8f68123cdbe024b"} +{"id": "train_042983", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "25f99390f3d84fca5fba73f52996c505f92cd922"} +{"id": "train_042984", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "422b585bf3445a03999e310be22f503e6cee5380"} +{"id": "train_042985", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f8cc82ed8243e1ef31fcf0fd36c8c13ecb062c9f"} +{"id": "train_042986", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2218137e33031fd666df43975307d92bb5612f42"} +{"id": "train_042987", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e32e1e7df6a52f62737d8ddd5a9f743bdb73251a"} +{"id": "train_042988", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a81f3a329a8b82283325a847dc727d3c2a25e7e6"} +{"id": "train_042989", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f36105498e31936f17a1f3b6ba47543c42bce675"} +{"id": "train_042990", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6cdbf07ab2690393bff11fb172a2d9d5783e08fc"} +{"id": "train_042991", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "645592ced75782e6fc2dfe2b0435ea8f4fc52761"} +{"id": "train_042992", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f5e8408d43b1599a7b99ee271c5a08aab0f12053"} +{"id": "train_042993", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "041c6c4ff8d37a8fabd03be8439828baaf2f5950"} +{"id": "train_042994", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a266a1a94fed777ed8d0210d76450bc60fd03e0b"} +{"id": "train_042995", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7236449ebe53b096eae2dd5a916818edb5ee05e0"} +{"id": "train_042996", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "baeba7d5bfbe388393beced0ffaa00c3b3cd61e6"} +{"id": "train_042997", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "720a711e8b2bd378e0a205dfe7977c061eb4ac38"} +{"id": "train_042998", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2d4770679b800bc593313b5ed7f28d20aa492228"} +{"id": "train_042999", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "b2c3e14685789c15aa76906e95d44a013828b089"} +{"id": "train_043000", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "224e9feccd9d53f9ee52389dab80241db8e322e2"} +{"id": "train_043001", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c2427bf736afeb4f98eb1653b6c2f43a785ff2f9"} +{"id": "train_043002", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "17b8226cd26ae7ea31e37c2adb746d9f0efa7d4b"} +{"id": "train_043003", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "11930b9fd87838c39fe15c14cdcf4f6a46f214fd"} +{"id": "train_043004", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7dbc5172a399311262a718319df63e718ca5b4da"} +{"id": "train_043005", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a42cb5e0ece5e60b073c4a8b287260d894fc3a36"} +{"id": "train_043006", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5e52c86c498077ead358f7b78263876381630690"} +{"id": "train_043007", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "fbf678bd4e86c841706d57cfdff8638f35dc105d"} +{"id": "train_043008", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "20f656c3d51fa71d36b1c1078c1728917572d20a"} +{"id": "train_043009", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7df70c0cf77a773f1ffb09309e0c6dec5c39b2f9"} +{"id": "train_043010", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9507e255e737133cd6cbd5baba07ecf9f587d23e"} +{"id": "train_043011", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "122e0cc1d8cd6fad6d34b75d5af713eeb5bdb45c"} +{"id": "train_043012", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "20ac7f1820778e8b7877adc65ebeac6dc23222ee"} +{"id": "train_043013", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "055df37c0b1a8a81ca52209a955e4076f7f1d77e"} +{"id": "train_043014", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ee3b1909ef5f07a8daff379ecd490dfb33fce1c8"} +{"id": "train_043015", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "463df88fee31a7ea9662826d5c670996a3d6ba65"} +{"id": "train_043016", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "718d79fd0863bff974d5518d5fe35fc4cfef6e64"} +{"id": "train_043017", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "99485c937a5629aff90d6ced9fa9848c76ddcb47"} +{"id": "train_043018", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3b2d2e8fe2a42346615994ca88763fa35f8549f0"} +{"id": "train_043019", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "4beada11c2131682bf2e0e32853b4f5e8eea8a80"} +{"id": "train_043020", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d20fc3392dffbcf112b63ef6029b6fe0e6eb7055"} +{"id": "train_043021", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "23a78313e39782e690daf518cb9d5fc2897b06b0"} +{"id": "train_043022", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "6d05dc7326cd030eb53633fbe80dbfa7ab81363d"} +{"id": "train_043023", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1ac3f35a57787d323c1e3df25b48f46d1f3686c5"} +{"id": "train_043024", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "01ff15273612e656ab2cd137e67ad7ca475da217"} +{"id": "train_043025", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "399737f25d00b0069931f06baebc0f742a9ae033"} +{"id": "train_043026", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "1cde996d8a4ac5f366b6978db318cecc440e0a2b"} +{"id": "train_043027", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "54eda9395a12fbe129276210cd728cb6b2bd0566"} +{"id": "train_043028", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "58c3ab781db2351ea813ba081b45b2aed2cfa525"} +{"id": "train_043029", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0c637d7c31f007706dcccf84bc2855a9afe3e18b"} +{"id": "train_043030", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8960e3e6d7a34013f9fe774b01b0c7076e947966"} +{"id": "train_043031", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c63612ad248a7a49fe8187cdf818cc8ddf2492fa"} +{"id": "train_043032", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1c3521ae0f3f137e8e918be9d0fedc98294d7aa5"} +{"id": "train_043033", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d511452825bb7959e59f9e1ac76540a1a2b2c062"} +{"id": "train_043034", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8c15b4a51cd92d8f922115318311ee338a0de4d0"} +{"id": "train_043035", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "06d197e731f20abb4f511b03c996b2b28afc73ee"} +{"id": "train_043036", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "33984b21492baf419a33a6d0b62cd684c7f6a85d"} +{"id": "train_043037", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "42e2570230da7d06d0b5713fefa827d15f60d3fe"} +{"id": "train_043038", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4a6e9726e1979c66c6ab4cd6d5c40e30e95ca6a6"} +{"id": "train_043039", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "7d6d10e5a94ab6bdf72786513389f67676ed80e4"} +{"id": "train_043040", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "32f1f29a7a96f7eb93a143d215ec82d5ebcff2df"} +{"id": "train_043041", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2bcb4caef884a0c35ff842bcc92d71fba52093b6"} +{"id": "train_043042", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "87f642107fecb78619b392702f06502974fd745d"} +{"id": "train_043043", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0247b493cdaf649323b959428413189fc9f36f2f"} +{"id": "train_043044", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "15a484160ea1382700567c9b3f738a62b5c860e6"} +{"id": "train_043045", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "6da05c87080caca3fb23c7aeb4666eaf3608096f"} +{"id": "train_043046", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8a22d1b96bbd5d8b9167dcc39f2dacfc50bf66a0"} +{"id": "train_043047", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2304fa5507f5e8fc3754dfeb47a5829ebcd14bbe"} +{"id": "train_043048", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "46a6d0f71216c2e7e22301bee3ae81de9706c8b7"} +{"id": "train_043049", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "3da06adc8ca459ba89dd01724aa858533e244661"} +{"id": "train_043050", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9dbfcfae9bc8a4ec87bfd47ea6495d3ac128b7fe"} +{"id": "train_043051", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "de24426a8b1db11dfaecff52c0965d4cc7b7b16e"} +{"id": "train_043052", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1add14247479cfdbb7c845fd31336c74191bb9ab"} +{"id": "train_043053", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6360d3a301d3145c116692b1fd6ed43310cd1d5d"} +{"id": "train_043054", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "52834b1f3eecaa1ff1f6044517b17679b4b97dd1"} +{"id": "train_043055", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f47f5a9992acce588141f75970e66c164cdfee96"} +{"id": "train_043056", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e46449e4a4df9b7c405455f95c6e1afa4b1461de"} +{"id": "train_043057", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "001b9390b2c3d857be4fca63ef9e297d4298e32b"} +{"id": "train_043058", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "507f766671e5c23b7e6e943281b60cf18e7ac780"} +{"id": "train_043059", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c227733b61672028886610a0362c4f68b2de7d09"} +{"id": "train_043060", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "53c01646f6a61c90c24d70b765d94ceadf889e24"} +{"id": "train_043061", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5ff8a55f5217e205ba33e2d57f77306407c8d0b4"} +{"id": "train_043062", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d7671d0ddd2a8b72022473f8254405378976179d"} +{"id": "train_043063", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "88fd610d4bbde333583c79a3aaedb6519717bce6"} +{"id": "train_043064", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0d2801cff61227ce6f96c63e1f9f9ea8ace971ab"} +{"id": "train_043065", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "368b5ee330914aa22a200a7a49ed42e0e4468b64"} +{"id": "train_043066", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "1bc2ca8386bd300096bba10f11a1f480ed24dfc1"} +{"id": "train_043067", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5b35aef15411ab36110703022c4672167f8e0198"} +{"id": "train_043068", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f8d226887e88aa48315ed64a82c03863ecc25ffd"} +{"id": "train_043069", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "49475817bde0b3bd7e408f99dc21debaf1ed6d50"} +{"id": "train_043070", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "fc91f3e21a61ee78057a852e93c3bfcc93728cbf"} +{"id": "train_043071", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e3416a1616d16d49ee337a45c642004fc4ec9f8f"} +{"id": "train_043072", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "06fabb38ef8a49bd6e82253cc35381622a596262"} +{"id": "train_043073", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "1c93f9ee85ce5090b9517aef02ef872b8e4a2a0b"} +{"id": "train_043074", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f41b28dcb95b3e4cc45e21bf9f902b41119582c3"} +{"id": "train_043075", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b5f4a07962dbb57b7292f3cb5d2051922f92d7fb"} +{"id": "train_043076", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "fa65d7cad06b336e5fc10244610ee69c2aac58c9"} +{"id": "train_043077", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2a3a05c6bec3e0562d38632a82ced67dde72fc42"} +{"id": "train_043078", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f1a44e12718a814fd8118b9bf1ef356b17aebc45"} +{"id": "train_043079", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bec18459a62dc511897633145509f8c33d7dfc8b"} +{"id": "train_043080", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3089ba7cf5ad8ef38245f2f62f8463c928ee699f"} +{"id": "train_043081", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a6588a053bfe7a14b3f359980987809c93cdf5b7"} +{"id": "train_043082", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ff51d2b938a7f8d0664e9aef9a9d9eb3c5625f17"} +{"id": "train_043083", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f5575fc6818ac86d66168760f255454f5f163fcc"} +{"id": "train_043084", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d738b1c35a8cc1effc0d6f7bb98f68649a201f6e"} +{"id": "train_043085", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "4adfcbe001b94dd4ba7074134cc829677124ccb5"} +{"id": "train_043086", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4d522a6f1903e59bdd8a6f681fdcf0afed7c7a2d"} +{"id": "train_043087", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9058e1cfcff60dc7267a5e6a743dbce9382b5904"} +{"id": "train_043088", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ddd759fcd5207081ffc4f05798c17b54ddfd5e0d"} +{"id": "train_043089", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5b9f48bba08fee478fa5266a201b12d86e329f5b"} +{"id": "train_043090", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5d02d4731008f93e996a3b97acb4ee4b1195675e"} +{"id": "train_043091", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "aa93deffeb1cd09f00208c50604c648aab8c76bd"} +{"id": "train_043092", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5db2b98f93cf3543e6246a813e83cacd1b4383bb"} +{"id": "train_043093", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2103ae3270b0675329b3f1b6271e7f107ead9099"} +{"id": "train_043094", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3508b133c073ea485757bb771e8928ae96508111"} +{"id": "train_043095", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a4f7971647fdb6cdbfd0526666ec2efb1576b053"} +{"id": "train_043096", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "08b2f3af640ecff2aa32b6b50aa00b307ddd7ba2"} +{"id": "train_043097", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8e4ceaeb168aae5ce18481d73af88ca054f98582"} +{"id": "train_043098", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2c0457be206f2799793c379702a8466dce0664d0"} +{"id": "train_043099", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9ec3b7835d76d1e361a5f00b5940eb2d8d692bbd"} +{"id": "train_043100", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c41482750e30f7ddde3c6e89554632a862295ef6"} +{"id": "train_043101", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "189e1489df6b93c0b84a5366e439ea734d4e67eb"} +{"id": "train_043102", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "abc3634d5f0d003cc43aa26e441733424265dfe4"} +{"id": "train_043103", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d2854a1760b5b39ef174104153e9574d7f68ce90"} +{"id": "train_043104", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "3bbc2c9ad1c93fb9a05901f34f540733611c8b3e"} +{"id": "train_043105", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "688510939442cbd347716f081b94ad4e42e7fd14"} +{"id": "train_043106", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "72bcb681f614f2f27998de4881972dedd96614a3"} +{"id": "train_043107", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "55ba7cb1bbf697f52807962c0df8ad0fae1d6002"} +{"id": "train_043108", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6bbf04771585ee1c630f615724e1d09a29f3a75e"} +{"id": "train_043109", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c37c6c924cc48bf2ffd173bce969263f9f599b0c"} +{"id": "train_043110", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "cd2f0bde2cfe9a3c01ebcc0cfe4ff3387b3cfd7d"} +{"id": "train_043111", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "3cdc405ffc4353eda05d3e210ce235de193ffc9a"} +{"id": "train_043112", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "586b683b08eedf933c0ffba1dfa38f643725ef7e"} +{"id": "train_043113", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "52284d9ee9b3f1b3f93454921c2fb0b942c22b71"} +{"id": "train_043114", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ccba1212b00654d7230cbf089603bb64256079c5"} +{"id": "train_043115", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0069a5241cbb5783eadd92bbd311ae93341c2b18"} +{"id": "train_043116", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0a5df3349485892b9a849ab2e7535bb59acf0ac6"} +{"id": "train_043117", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9e31546dbed7613885ed00441615b0e4ba55b0aa"} +{"id": "train_043118", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "0b68b96b439fd6f2af6a11bc3400bfc56664d7c6"} +{"id": "train_043119", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d8f37a5e60f037147c4bf6f5e35e25f10a8c1f35"} +{"id": "train_043120", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "80828d3b771ee18f8edd5c153eb47607651ac844"} +{"id": "train_043121", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "950aeb798e72165b19dd92dcece0466eebbf7297"} +{"id": "train_043122", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a57e7c2a6c62a60a20bb604d6ef6f1270c2e2316"} +{"id": "train_043123", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a17b4b64bdd8ddbe1408330b43a3ab59db74dbe0"} +{"id": "train_043124", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7ff531d476b03b47f275d5656031e10fc5b1b6d2"} +{"id": "train_043125", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "36c70049968ac00fd297e4eff25f0ef4e832be61"} +{"id": "train_043126", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "24fef6ad33b006d7ba0f8568b847ee18bcf520ec"} +{"id": "train_043127", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "74ac7b7020f97679a416bc3c9ce50a7c54010471"} +{"id": "train_043128", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ebf9703928c7c6ed76da7f7beb3a982246825c3e"} +{"id": "train_043129", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d9f377d764e6c5d50041727bd61f903d816184f7"} +{"id": "train_043130", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b599cf079314b0f278592a7e9a726cc31934ca64"} +{"id": "train_043131", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b2924b715989dcea6af5d0240d3e786089e6f6b2"} +{"id": "train_043132", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "420737bb767358fb741869f8563ab176cd4b87de"} +{"id": "train_043133", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d3b475afc89710a62ed6beef3a796994904015a4"} +{"id": "train_043134", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e3760cc363be39415e22d2b593a2a79deedd91d8"} +{"id": "train_043135", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "28b6c4aab232933797a96e08995b2e0e6f175766"} +{"id": "train_043136", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4d2fb8773000378a47c7220518770aef09bea3b3"} +{"id": "train_043137", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "016ab9ffafecbab94afa1f2f12049015e179b480"} +{"id": "train_043138", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6427feeabae03f1ccedb882f4b980431369c577d"} +{"id": "train_043139", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "462184157ae2fefa9a2c74d6141e8496d629e4f2"} +{"id": "train_043140", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "64a655ffbf298d843456b63f7bf1c3216d057de9"} +{"id": "train_043141", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a360514940f404968d4f8d7614fa44f58764b4b3"} +{"id": "train_043142", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "9930b2258527a1f9e171e008d0f7946bb08cfd33"} +{"id": "train_043143", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c47326ee0aad2508c650afdd0165cf59e260b6cf"} +{"id": "train_043144", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3ac648631d39f9b6278e82a1c8d200f3cb906dc5"} +{"id": "train_043145", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0c3602287f179470fd9b5b00a748a671801f7896"} +{"id": "train_043146", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "eae8875621837df62953057dfddf9dfeaa05216a"} +{"id": "train_043147", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f7e85c736bbda55208f564a227a99433ce48cd42"} +{"id": "train_043148", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f2e4ba6e1a48211d75369e9668d6d56ae31ec682"} +{"id": "train_043149", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "85c2af5a32a94e39ca064332e87d44db6c8d11f7"} +{"id": "train_043150", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "15fcc40fffb56450f794f3dcf43a9ea486cebe8b"} +{"id": "train_043151", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "150aaa4e0f7c363014f6c10c88f3b5a4b5b35fbe"} +{"id": "train_043152", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "698804ee6110287cfdff4bee04787388b80ad193"} +{"id": "train_043153", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "7e40ba836d511862b33d88467104c63b33ceb1d0"} +{"id": "train_043154", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "06941dc6f66334fd97678ce761009afd852ab9d3"} +{"id": "train_043155", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "366c667e83e68564812c987411a7ee8157d414fe"} +{"id": "train_043156", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "b51e19e2cfd087f48caf6644fa410f47c92410c2"} +{"id": "train_043157", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1f512a98a49d700e942b31ad64427344333b9704"} +{"id": "train_043158", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ae6ae04ce9ea3dd46fb74aaa7206654aab30bb42"} +{"id": "train_043159", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9b2c8945e50250e91810116a85b7d20ee63aa7a4"} +{"id": "train_043160", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6907e7514f933b7b5b07ada21836912cca4dbecf"} +{"id": "train_043161", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "20c132e3beb119b1d2ab68039f03c8d3cd97a862"} +{"id": "train_043162", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0f29c651bccad8087515c07e1271d77524f01f54"} +{"id": "train_043163", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4c960e2b387c35c26eff82016a33f796680c2b85"} +{"id": "train_043164", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2791f665838fae460710e97cc0fd6afb2974360f"} +{"id": "train_043165", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "975fe26be668ed6d224566dd9057e4d812467e23"} +{"id": "train_043166", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1f958ab19a16c953e8fdf5b7c3fc6b3671e68d70"} +{"id": "train_043167", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3fcd2f08e097fd36735bb9ff6ca676331eb4e135"} +{"id": "train_043168", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ff0d96bb8b3fead610cb00b6cad2b0e28c95396a"} +{"id": "train_043169", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2ef3748285da8790cd5709f6c6e0ccb4745f1349"} +{"id": "train_043170", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d26f854fb8dc6d3a7e3950e4fbfe764f379c5806"} +{"id": "train_043171", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5c75cf9474888fb61bb189beb14f555afb1c4137"} +{"id": "train_043172", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4112a0fbbd574836c8b549ceef67af59da9bb79d"} +{"id": "train_043173", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0440d2ec404d85b1c620e61cbb013281880c4486"} +{"id": "train_043174", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "58637139f076ccf715371fbe9982b7e9bfbb6892"} +{"id": "train_043175", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "404c55f8b985c5d8dbb19d82aa9555500ee564f2"} +{"id": "train_043176", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "819e4932fca50f69c8bfa3cb0d03438cfb45fc07"} +{"id": "train_043177", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "85fc53b09d7af8c09567edc87b40621a9f087722"} +{"id": "train_043178", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "061fcfe72aee4f01f845f0ab731895b0253846b2"} +{"id": "train_043179", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a99062cc419d9b5a0c44f59914cfe70eded945a6"} +{"id": "train_043180", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c3c6760d31ed728b5302731c62766946ecd831f5"} +{"id": "train_043181", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2913d213c5042719f0619a0abf679c9aefc2a4c8"} +{"id": "train_043182", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d08d148cde4f2bc6a402ab06537cceb59c98f0dc"} +{"id": "train_043183", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "dddbe85d0ddd19506ddfc09d94e42deaafc60682"} +{"id": "train_043184", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "73b7a2d948067281e77bf0e75d36f49594c7d245"} +{"id": "train_043185", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "727514e17ac581fb2f23e9084030341aea143873"} +{"id": "train_043186", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "acae3e828a78da1aea95049579de909d907f4094"} +{"id": "train_043187", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fcef62755b94b13d23595b65e24e70eda0457dd3"} +{"id": "train_043188", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "beb22ba1130b16542e18cdd6116db4b3f20c2bac"} +{"id": "train_043189", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9004dd994be2c04e08d128fc1d7233eeccd775ba"} +{"id": "train_043190", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "13121131b075cf58226ed1c3ef728dcaba6e154c"} +{"id": "train_043191", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0073d9f517528c565ec725ac49f21a6e11dd687d"} +{"id": "train_043192", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f49414a44f032bbde064db94211ba3795be7a659"} +{"id": "train_043193", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ebeee5977408824882e8458815de347644f5d310"} +{"id": "train_043194", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5a0ea2044e41ed1c87d692e634593a166e494b11"} +{"id": "train_043195", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c5858e84e1088c06ed5ed6ed3e7a9c8552e9da82"} +{"id": "train_043196", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "45fad5d69c681df99342f89dd1cbf9c5208608cb"} +{"id": "train_043197", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "11c4278893cfac78edc4eca69ad21425c69be431"} +{"id": "train_043198", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "33485a071d07bc780c6704efe6bcea6183ad80db"} +{"id": "train_043199", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "072d64ce27be2199514d04771c5a1caeebdb22ef"} +{"id": "train_043200", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "fe19ca7af37f3017f12ac87226cad68e1fd046ce"} +{"id": "train_043201", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "508a68a32c018907ef856929c644577a8485b631"} +{"id": "train_043202", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "eb5d8a1bf8e2e88684b8c57fc1c48d00ac185fca"} +{"id": "train_043203", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f9487609826af0822eebb6a84273c6590fcdf23f"} +{"id": "train_043204", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "14da01080daf8025c57a661a26da2de12d9167e5"} +{"id": "train_043205", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1e43a9eff46635a2597ac611010e182e591d52cf"} +{"id": "train_043206", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0a6840d4ddad75e0d364d52255cde21f60850a72"} +{"id": "train_043207", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5dacefd5effc4db10a7aa7c34abcad78f790d236"} +{"id": "train_043208", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5860f0482d4befde9b19e262b615c1d7d60cffd5"} +{"id": "train_043209", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0413991dc831166300c2616650ae0081f5230756"} +{"id": "train_043210", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a388425dc7efaf00b2d9c99ec2df760773074491"} +{"id": "train_043211", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "00de19cc09770f2579c1c7ed38beca6c22750754"} +{"id": "train_043212", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "915afac4832676a45d9ba2394b15a18254252183"} +{"id": "train_043213", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "166ede37be43d45963727a43a85fb502dadf6f32"} +{"id": "train_043214", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a07f1a2f16fd622aed357a8b8e20a81b66f98b4b"} +{"id": "train_043215", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "571dd555a0d8bc31e8ac159c89f543eaa93ed348"} +{"id": "train_043216", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "c4dfbe86124ee357d51a684c77d4ef4d35d70171"} +{"id": "train_043217", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "fbebf3c0db9870560b681eb362112ccb89b97024"} +{"id": "train_043218", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fb19854d1e4cbd11a5b9c8c329c29d6f28f56c13"} +{"id": "train_043219", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "fb41125ff1c750bcdaac8d542a050d35ef384286"} +{"id": "train_043220", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e9c9f410bd88d59ff89ad8c4830e146e3ec3ab83"} +{"id": "train_043221", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "58274e7ceda56ec6f9c5c0b375baafaa374ed234"} +{"id": "train_043222", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9d717bf0172102a31e9d2e711abd6d5293b67e77"} +{"id": "train_043223", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "88f514e4cdddbb9177232ab4209b22206a476266"} +{"id": "train_043224", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5f60bd1f9b8ded5303eec07d0230d2372510b375"} +{"id": "train_043225", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0ca87ae4272f6a4f32d5db207cc8f943f4b1ca2d"} +{"id": "train_043226", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "162e29e2a6d4d16c19d93c3b5422a7a556542ca8"} +{"id": "train_043227", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0b880bd0e674d53ac88805a035c631e9a099a928"} +{"id": "train_043228", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "421cc5219e96f86f51e349fade9b7b6d4d5756bf"} +{"id": "train_043229", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e87c9aaba1e91ce3f639f1ee38f9dca1a285cb92"} +{"id": "train_043230", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "688bd9c277ebe8c1506ca5d8c3c035c8d09fc55c"} +{"id": "train_043231", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8571463c4071e7a885e85afdb2badc15de5d722d"} +{"id": "train_043232", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "1c491e647736e14da5e86818a659b342fd320263"} +{"id": "train_043233", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7b88774c246a8d08d8ccdfa222f282c786f0108c"} +{"id": "train_043234", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d97fac731fb20aea8e8d3aeff5bbe37e825a599d"} +{"id": "train_043235", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "869439015bfaa1cb4fb2d4d09ab3b7855bf471c0"} +{"id": "train_043236", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "345d0a6d2bf87de2dac5e27925e0eae582cacef9"} +{"id": "train_043237", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "eb2d9385f8be676c4299b76f2effc90c0d8c6602"} +{"id": "train_043238", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a00f0f3751b33732e49317bdbaaf9aca07f259fc"} +{"id": "train_043239", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5694651f444b66fd8f1dc2592d7f5940e58bdd6f"} +{"id": "train_043240", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8de538fbbd162468d6224ffc85942afeb5090e69"} +{"id": "train_043241", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b25ce8833636bd9f62fb1d17b68576297ecbce46"} +{"id": "train_043242", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "de24426a8b1db11dfaecff52c0965d4cc7b7b16e"} +{"id": "train_043243", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "08eef8dedeb2e60b7bbea5005af36693e5f55a38"} +{"id": "train_043244", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e0100b1a6c246288a06a170f332ba469d1e985ca"} +{"id": "train_043245", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "e67e7493ff9ac916ac3f096cb44d3a0a70b20d28"} +{"id": "train_043246", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9b2415b173755d0e30b9398f5b9657f236b77e97"} +{"id": "train_043247", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "124db14f34298ff0c5c36488fc797b506fc072ef"} +{"id": "train_043248", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e1ab0cd71d20280fbfb7659ce9d52edf93fd6390"} +{"id": "train_043249", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "728a738077677a1fbf00c3e7a6933f953c1c1825"} +{"id": "train_043250", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "80d2cd8d2eab669084b7dac416de1f42780c475c"} +{"id": "train_043251", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e0a9e5037d05679fceb5f346f7af48be2eacc118"} +{"id": "train_043252", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "fb85c5617c14006522ef5a6e36ee6430d93e9f10"} +{"id": "train_043253", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ca498854a5f5c45d971b28c1a2c963a86c2660cc"} +{"id": "train_043254", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "6662f80197feb5f56573ff68064bccf6b4fb7417"} +{"id": "train_043255", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0530590d2195228d9f6ec9c7198fddfaa61e68cf"} +{"id": "train_043256", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cf3c877eba8979262da591a7e4bae64bf7a74d9c"} +{"id": "train_043257", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fe19ca7af37f3017f12ac87226cad68e1fd046ce"} +{"id": "train_043258", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "78b99b1f3d028c65a99c672425c103295f49bb96"} +{"id": "train_043259", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "0ba5c3024e6eb7f4bcac7c3c32480ccde5566a00"} +{"id": "train_043260", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "eaeca5c1d241349cec1fdafc0964bc577e35ba06"} +{"id": "train_043261", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c87c3c591b623eeef946540f973892d5bc16f3da"} +{"id": "train_043262", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c844928378da5ac214dd3194ee50d1a1a3285d39"} +{"id": "train_043263", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "441286549af551f1a1ed1099af5178b059b2c047"} +{"id": "train_043264", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "083e3f72d686f60e0b1437e5f45678503ae39be7"} +{"id": "train_043265", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5c7fa5c83030db25736f19e892c794d369a4a3ea"} +{"id": "train_043266", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cc043ae63e8933e4e7a84920aa97424639314748"} +{"id": "train_043267", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9ba78f53ae2dbe6117e193e9edf2e1f25e41ec63"} +{"id": "train_043268", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "306b0345898e9249f7262c0b53d037803672c28e"} +{"id": "train_043269", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c375d9d8b8cd1bea89d48a9547fd61975d84ba36"} +{"id": "train_043270", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fa0dbd5f7c8cd011afa50a44e0bfa9411189dd25"} +{"id": "train_043271", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "bd09b51e922610eff93548d1c73f1f0506416823"} +{"id": "train_043272", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8cf30d80328ced0cdb770fe66a41ab1fdfe97439"} +{"id": "train_043273", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4c0a7cc444d049c7dfa4454341dfc0e8b42d6ed0"} +{"id": "train_043274", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "404f0bb8aeba66a014c3060fef83dcd550a33b23"} +{"id": "train_043275", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "925beca824ca739536c5bec0973ee36429432041"} +{"id": "train_043276", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e6ab091b50a564ae541f3ec28cb122b4dcb9f8d8"} +{"id": "train_043277", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b8e0c259ffa6dae9076806ed35691c3435fa2b7e"} +{"id": "train_043278", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3149f56b2ab271a091a2d66338f317a9b50362c2"} +{"id": "train_043279", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e4d74d17d3abe8031766bf607feabde9001fd929"} +{"id": "train_043280", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8809ee7a253ed82ee488bc84799d5e5efd10d5ac"} +{"id": "train_043281", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "bd7a34d47c94cd17cc59fbad338c0cb514860607"} +{"id": "train_043282", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "db53ef86558bc20a4fcbd8cc4f898e67e915512e"} +{"id": "train_043283", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c12e00c4ed89d3fa406b98495458201614d08d7e"} +{"id": "train_043284", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a3f7950a8dc33b5103121d1fc5f8cdcd2a555c68"} +{"id": "train_043285", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "bdb8f3e3a4f763b8d85ff5a0bd16ac1f622dd390"} +{"id": "train_043286", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "05f0016f264396f0190e555460d9fec2e23f6700"} +{"id": "train_043287", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "52ec5173f7d7ae6863e3b9902779695643ed7f38"} +{"id": "train_043288", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "6a40986fa7437730ab0833ead438b06d4aab3cbc"} +{"id": "train_043289", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2cfd709b5f0b7878635daedd111e50fa63a2bc2a"} +{"id": "train_043290", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "7cf34dda10f91500c236991bb97f66ce451d244d"} +{"id": "train_043291", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5872c5fbf902c4e5694218985bfff5d07d5c9af1"} +{"id": "train_043292", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7c8f21470d5740754f01f25f82b09517ebf565f5"} +{"id": "train_043293", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "479f031d26f8c0582f8c0da62ff56203f1cbb7ee"} +{"id": "train_043294", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "182453fb424eb57c13ed93b336fa828d29397b0c"} +{"id": "train_043295", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f707ce89dd15067ba44c455c4291ac91143fb3f4"} +{"id": "train_043296", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "5f6c0374521d5dc423769891446dcfa0f4590f24"} +{"id": "train_043297", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f264bae0a39cef504b31915d9871615e76755f48"} +{"id": "train_043298", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "476ff383d16244219dac072430c788530ebff45a"} +{"id": "train_043299", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "d471e15d4761dd6f46930cce85f3c825cb2772f9"} +{"id": "train_043300", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6bef381cf87530b692dcdc20da28f63ec966da4d"} +{"id": "train_043301", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "998ad25cb23c5117a15befbc7a3b13b5d2be3ade"} +{"id": "train_043302", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "405c03de446997c76b81d3f0aab89e4d371e5403"} +{"id": "train_043303", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e0ff3a7a273ceeb883099a9701245cedfaad9176"} +{"id": "train_043304", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8e0273ec161d7277b15db018b88aefa333297a20"} +{"id": "train_043305", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "eab556d42943bed383a13b09781870200bfd5301"} +{"id": "train_043306", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5592fe746dcaca494bcbe15096151f13e652cc2d"} +{"id": "train_043307", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6ad51610e8019d069ccc33357b1360dbfb498887"} +{"id": "train_043308", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "cef98ee25a9fc7c40cad115659dd4e4bd29e6661"} +{"id": "train_043309", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "35d1276f3737bb3a75c8deb4b07cbd82bfe24566"} +{"id": "train_043310", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "09ab1074f3e9784b40cdca8b6b56808cc5ac4aff"} +{"id": "train_043311", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "84057f140a37aee2b4e8ad7ededa7c83597b26df"} +{"id": "train_043312", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6cdcefbe108298c3adf26fcab01d1bbaaac67493"} +{"id": "train_043313", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e302b1fe904a059294e575177aec11fadc3d65ea"} +{"id": "train_043314", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "5e03356b68cbb94eb2c5c9d6bb87ea4fe081e802"} +{"id": "train_043315", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "56cea3ca8e5cea70e4ad69a85c908b19f0076c28"} +{"id": "train_043316", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c4094d30bbe2568e4af07cc86f2cd1e216ebccb2"} +{"id": "train_043317", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "c3a51b37443d8dbb3cf0219cf40de48b80339f74"} +{"id": "train_043318", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "56a0059adc06dc0afd3bce6e0bbf9d21381c1cc0"} +{"id": "train_043319", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "dceb52a6468884a7ceff520ad1775ca481484acd"} +{"id": "train_043320", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c0d01b66648475fadac521f86cca9840f9d8e8e8"} +{"id": "train_043321", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7ab5d25ceacea14c2bdf4bda47f0a2bd1230d006"} +{"id": "train_043322", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8290af07c4dab78e1eed6406a3831e8f48859118"} +{"id": "train_043323", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5c9ad105a27ae3c9facc756a299bf27740fbb8b7"} +{"id": "train_043324", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "53505dcdc76f832eabf6f54051c1d500532ea24f"} +{"id": "train_043325", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4dc9878093daf2956d3145ac4cd671809731651b"} +{"id": "train_043326", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "43c3e2dbec0d26c38b59f3e1d231bbdcc9efeeb5"} +{"id": "train_043327", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c126946c4b0e0009e975b2834ea1afb1f97c1104"} +{"id": "train_043328", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "3d3ff2e39842660ff24df67bba55ab17890b05c0"} +{"id": "train_043329", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "14a80e0fcad8fba7a49324c3ae39f0fffccb5294"} +{"id": "train_043330", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "4c8e19e2601ae49291bc3ae9d207bfcfa7faa60e"} +{"id": "train_043331", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d5cad8ccdd4c5ec2f7b7a67c45cd21174f19d207"} +{"id": "train_043332", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f5d0fd588dcfd2bcab65e4b8c474e419d21d0242"} +{"id": "train_043333", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "00176edef004ffa2eb91253efd9adde9fa7893f2"} +{"id": "train_043334", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5eae9a2a5e984922a42b65aa92e46384f954053f"} +{"id": "train_043335", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "fa187b39eb20deb39785fe15808cc50a4d38efec"} +{"id": "train_043336", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "572341656f4fafa753852753584b4ac2bb8f221c"} +{"id": "train_043337", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "648a85ab7b8a6a189e3e684fe1eaf4957dfde675"} +{"id": "train_043338", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "40d36568cd5b823bad35b76eef7d9af7b3b94ea8"} +{"id": "train_043339", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "27aaad0cd7a8f0819b3e5842a388798c5fff5c39"} +{"id": "train_043340", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "73401425426a5fdbd5828246a897a58b72daf6c5"} +{"id": "train_043341", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8b5a1278bf82c516aea88a356b2c4d6a08033489"} +{"id": "train_043342", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "71894cd5a36997af87b8ea42bbc288814c2b68c2"} +{"id": "train_043343", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "4a6eb848575bc59213b9a48ce86f9c96dd995a7b"} +{"id": "train_043344", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "75bc660e35f6fcaa46730fb1745dc6065b313ad8"} +{"id": "train_043345", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "38c547184f396e1219360e5e04e16122a1578f54"} +{"id": "train_043346", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d23dfa18e63f920cc87f60ca915f949575efdde1"} +{"id": "train_043347", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8e0ff9c5e97003f04528f76457ee2eeed863ae38"} +{"id": "train_043348", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c6e8616d79573bddac1e224cc84dd1a513f1a749"} +{"id": "train_043349", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "104a5b3a6d07fa556f71d70476d93b317ad83184"} +{"id": "train_043350", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "cb7a74a8ff069a60fccfb0c2f9ff1d1d04e7f4a8"} +{"id": "train_043351", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0fc77a4db69f987b4fdf659b97e7b5c6ffcb955c"} +{"id": "train_043352", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a537ee2978e640c477bd26bee97abc671729c521"} +{"id": "train_043353", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d805d831cd9bee0f54c29f93e46d8f457147ee1d"} +{"id": "train_043354", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "67845c4bc98ba0a5a0295a06198ad6c49eef575e"} +{"id": "train_043355", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c4457d9efe5e282ae3f5b23f454b69b72eb79543"} +{"id": "train_043356", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3a24cd17276b4c5e4de695ce216f4e4ba34eeaa7"} +{"id": "train_043357", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e2483a66dcb7f1059b662f2d065f2ebeea042930"} +{"id": "train_043358", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3001d8bad6860bb5cd502086889f7d3a319ab5e7"} +{"id": "train_043359", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "78cf6addfdc5256989cce4018819ee7321437a39"} +{"id": "train_043360", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3a4fa03e7e1c5da900cfc4ac0ae51ae763a0af86"} +{"id": "train_043361", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f3e97f36bf66e896275a14db80c947c5c9af011f"} +{"id": "train_043362", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "361050e77b8aa6e9614e86b51bbd7a5a5ce8ee0e"} +{"id": "train_043363", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f65c44f0d7a8cd8fb5348a2ac6e6e0d3fe01208c"} +{"id": "train_043364", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "bedd6e02b56707cfc25500b04dd631c76f4d80f6"} +{"id": "train_043365", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "81ac0107eb64ca6dfc7c1ccaca92e66e2eae4cb5"} +{"id": "train_043366", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5f199d41bf646e391644cf2549e9091938f721e8"} +{"id": "train_043367", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6382f18dc1545604254c661f9074a23b7c849d45"} +{"id": "train_043368", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5248d7bd20bedc7f3e284c559b860ffd848792c4"} +{"id": "train_043369", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "94f053f104d9ea4bcb218443da1f153359e44b94"} +{"id": "train_043370", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ad6a135469cd0b0cde64e74aa8c39c414e232245"} +{"id": "train_043371", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f76d8396975edfa5f862eaca700ff3499349ada2"} +{"id": "train_043372", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "357a7ee954cc888d18082a1da8a80cc852fbaab7"} +{"id": "train_043373", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7693806cfe880f9ace1e9423d5330eaa5a63934c"} +{"id": "train_043374", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "475aa5af21782796025d7d2581e87ff8bb349b27"} +{"id": "train_043375", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c1969ba3826379e5a9465077d854fd490e3a68dd"} +{"id": "train_043376", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5094eb5ea2372f382d06f787f2ecc8bb7de8cc53"} +{"id": "train_043377", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "290da7e2caadd9c61617c64348a3d0b280715e32"} +{"id": "train_043378", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a42b138983a384ea4ef9327b4b687732eb2b9440"} +{"id": "train_043379", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0369319836a6ce66d48c21129f7f4012fd78464a"} +{"id": "train_043380", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "716f394920728c66079f43a6a420884b1045052d"} +{"id": "train_043381", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "33d5f9408e54afcab24561ab7f3366303e0f03a2"} +{"id": "train_043382", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4beada11c2131682bf2e0e32853b4f5e8eea8a80"} +{"id": "train_043383", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "abc5f2b11ca7e71cd6a71bfdd8d0af57c814dd9a"} +{"id": "train_043384", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "64d591c19c3cc678791fc7c237625cbc4707963c"} +{"id": "train_043385", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "1257341cbe26cdf5a9bb74498477edadac1778e6"} +{"id": "train_043386", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c01dc841cfe96452c1eaef7c07fb34750f334521"} +{"id": "train_043387", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c936d12a5280bc6dd6af681c597fcc485e40ce80"} +{"id": "train_043388", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f19b3b3c5d68087f81879a81ec706dd0d7d14f63"} +{"id": "train_043389", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1599889f21310b4fe8434041a4babf9a72238395"} +{"id": "train_043390", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ef01cfb40d6c912ddbd0ab3f4a02d94155f48f59"} +{"id": "train_043391", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2dc9388e3b9ecab34bc12f0758ab172bef77687b"} +{"id": "train_043392", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c3575b4d6b302f14cdfff28f4987e99c9bb564e7"} +{"id": "train_043393", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "fefd1408d4ac60ce760d65e5decb4fd0faf714e2"} +{"id": "train_043394", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "addd97025bd6f1427c14d912b18e73421e587424"} +{"id": "train_043395", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ca8bc75aba2d5e72437db5e3c090111f57f9023b"} +{"id": "train_043396", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b5225cf147ab737204c80562304e9c9293861569"} +{"id": "train_043397", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "348fe4e0e893334d269e3e4c9adf647b24814766"} +{"id": "train_043398", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4d1a2b6648752d621114ccd4e5e45646053468aa"} +{"id": "train_043399", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0f4b7f8c0daacc1d991863b1f8feceaaded2b099"} +{"id": "train_043400", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f22f06fad751d7952d8498a6601fd4434bcef51e"} +{"id": "train_043401", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "08bfd42abdae72d0302cb9eb0d7bca16b5edaeb3"} +{"id": "train_043402", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0f99de0fb16a3197cbebf0b287b8efda19adca98"} +{"id": "train_043403", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5d99586bc06bd81208c069c417999dc173645e1d"} +{"id": "train_043404", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "94b524dd95a1bae0d9ec921939e770356e51a734"} +{"id": "train_043405", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "89c9233d449e27829890261f5c5f6f0475bd12a9"} +{"id": "train_043406", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "066c7be4bd6ebb68653a66bd1a11324d56ccdc8c"} +{"id": "train_043407", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a90fed3ec16eab870bb52df7ea4db8eac8c43e0a"} +{"id": "train_043408", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1e368309e3314dbd70a8fa30e60552c4282b52ef"} +{"id": "train_043409", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "19ad38db1e1666b4fded635224998dfa5afe999f"} +{"id": "train_043410", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c81ecfe17c0907b2e620c97b16f01327291ad7f3"} +{"id": "train_043411", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d48b537968ca4da9924f0ff8df7560bf8a17e0b6"} +{"id": "train_043412", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f4be643f6ed2f66456ba0a593fef794fd3e0cda3"} +{"id": "train_043413", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "034c01dab8c1f7744b804c5b0fffdbd74967cd41"} +{"id": "train_043414", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "003fded837a95248f4ac57c64ec4a579ef47ba4b"} +{"id": "train_043415", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ea909ac8e12a13c2b20cfedbf2d85a914a14c8b9"} +{"id": "train_043416", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f21bbed606fc8fad5da607da224a51787039ef75"} +{"id": "train_043417", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a6abfb8f4225a3ac57f32bbda1a802280588ad90"} +{"id": "train_043418", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b44f81323d2cdc66e23b1e6c449d416457842f04"} +{"id": "train_043419", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "91fbb42b2091872eb47ef6665dc201c4a5b8e801"} +{"id": "train_043420", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "681eb3fae676401c84592460bbe48034b84f6291"} +{"id": "train_043421", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "da187dc434fad34d99ea81122701cbaeabb2e175"} +{"id": "train_043422", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3097f136ffc082d50db11dd06d7f04125283cd73"} +{"id": "train_043423", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b6b4482644dbbec0272dda8dfa4168245d1c3e17"} +{"id": "train_043424", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "59f2840b0584abdfa6f28b7447fe7fe2451e4491"} +{"id": "train_043425", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b2638c505e2c3a76674bfd2185cc60e5c175a149"} +{"id": "train_043426", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1850c90da7c75f5d0958bfecadde00e7474e52e0"} +{"id": "train_043427", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "728ffb60f5f5697d1f0ef3b98b5f21762cb4b014"} +{"id": "train_043428", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5f39e7c1ed318266a8f9a65d2208a95c7b240879"} +{"id": "train_043429", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "de61764b03b02bb557e59aedd63c09d3c9c674bb"} +{"id": "train_043430", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c188693c899b382bf89f2e9e549c87f294d34d09"} +{"id": "train_043431", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d04b0cb2ef777ce2a2c44ba80c4b725827ed6254"} +{"id": "train_043432", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9d7196f91776b4a14da930f7d5073feab8eceb8d"} +{"id": "train_043433", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9812b90823ab73b1ddbc92246d2b2caf4887c3d7"} +{"id": "train_043434", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dbd96fc5177502ff8a466f50b1c613b065b23b35"} +{"id": "train_043435", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "26410ecbfa4cbcb13396d0a17e478af0de859fa1"} +{"id": "train_043436", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "12f101023cf07de1f24c5f37509c4466842c9476"} +{"id": "train_043437", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1ca4cdc8f6c576cc90db5e5ac5cf11f1f66ed530"} +{"id": "train_043438", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f082ff81a8456cf279ede37563b6f846e0acb549"} +{"id": "train_043439", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "46ccc56a958374dfc44472ff61aa09a32b432a39"} +{"id": "train_043440", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "cf8a1a8e1ee17f230cf53fcaae829480dd656641"} +{"id": "train_043441", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c85f6326d644e1cd631a5c1a6e180ad160a24729"} +{"id": "train_043442", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9143af4c489cc91732db0178ee90d0fbde5a9b2c"} +{"id": "train_043443", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9bb75fe6021f02266f1aec5763acb97894d0f286"} +{"id": "train_043444", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9a840204211ce7fbfa1916203becf8a9632949b3"} +{"id": "train_043445", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ca4d6e33bf8e3c66d9dc9663f42e85f428cc21c3"} +{"id": "train_043446", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c012639704990f535e8727240b5876451b403b38"} +{"id": "train_043447", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6f5419bb6b4c7613d8130cdd52dcc08c122c27b6"} +{"id": "train_043448", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "5330b3e7745fd0560817c1639138f75b9f774599"} +{"id": "train_043449", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8d137bd5f20cfc94286b806584d386e033b00d18"} +{"id": "train_043450", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9d544d219fdb3d761c49e545dd32410b4edce654"} +{"id": "train_043451", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7e9ac79c468ae04ea5c695e2945da60cf82df0de"} +{"id": "train_043452", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b2a01d1cf5c9c0d4776d50464098e9919ee9bd58"} +{"id": "train_043453", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "531469944fca36a8c88e3b5e754b6cff8967b341"} +{"id": "train_043454", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "bb5b38717670d43ab9d781b0a930afc639748eea"} +{"id": "train_043455", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f652c2d9b8e36e4689da2c87f6a7386bd4d86a47"} +{"id": "train_043456", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "97fb0165c60f96a5519a6ce3150e88bccd24d5d1"} +{"id": "train_043457", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e2bc21b61367e87e24c64df0ce80b8ab165ff2b3"} +{"id": "train_043458", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "078d988d77765103385f65cfbeddabff2871da88"} +{"id": "train_043459", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "de6745ccb479d6d8f158817aa1359734981c095d"} +{"id": "train_043460", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "565507fd2c3791aad5e5498031e257aa37386bf7"} +{"id": "train_043461", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "99fdd164420e5eb2c4cef9b44ada8181ccbc6a3f"} +{"id": "train_043462", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "63f883661575f9a44aff1bc65c29f09879c8cf3d"} +{"id": "train_043463", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "c79170fa06cedd25a201bca12d14e3ed85389bab"} +{"id": "train_043464", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "310e6abb902124f070b5e8560049c910e21fc023"} +{"id": "train_043465", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "969d832e43a71ec58e885144ff560db45ee5fafb"} +{"id": "train_043466", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "4e89cce436bf088e2befe1ed4aca8ae172d360ab"} +{"id": "train_043467", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "1867f302c11c8f8836651700def89b27a0494a48"} +{"id": "train_043468", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "38d52395a460afea6be2d2064b28791c68259aba"} +{"id": "train_043469", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a3896b9c20ed1e1d53fb6ba6bd56bb4886819a3e"} +{"id": "train_043470", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8568ffa69749b6adaa2490970d233c80566eb883"} +{"id": "train_043471", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "99ebf4cd96a94b29b0af2ab61e15b2e604cb92f0"} +{"id": "train_043472", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f00920b7aae7180356f3e671fbc2ca4e1f5c3bcc"} +{"id": "train_043473", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f851f5319165c8b9c0d3b4504d6a41f89daea87b"} +{"id": "train_043474", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "03daaf7e52f400ebff8481878c7a5312c956bd75"} +{"id": "train_043475", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "462d9b154e12a6508aa1008ab3e2a4fd375fbcfb"} +{"id": "train_043476", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8dbd9a251137edfc333b6b0303f1fc884ff485f5"} +{"id": "train_043477", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "05809682a5e168fa4892f81388e7f11026e4575c"} +{"id": "train_043478", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "bcce56e3f0384c2815c52d432dfbb7893db2db8a"} +{"id": "train_043479", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e40a9b3f85dd63be6d687035c9727e82b369474e"} +{"id": "train_043480", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "cce43c5c22f5db098d8aad85751789e097f2ee90"} +{"id": "train_043481", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "292c3e3035c2cb5ed8f93306cbe6895da0acb289"} +{"id": "train_043482", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8a45c046a31a15052d95e187662198b7fef6ca04"} +{"id": "train_043483", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d03118fa8aad4477dcc6cc06e31e6a1af68e580b"} +{"id": "train_043484", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "eb8f6c2df190283693d107b9eeb30465330ec2f1"} +{"id": "train_043485", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c951e5c2ecb0e21f751fd9bb21331ab52eaf61c9"} +{"id": "train_043486", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c2b6efefe1e24aab94f959f966f306c57971a661"} +{"id": "train_043487", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "add960a85c0b3b2de909503cf30ab05faf83f507"} +{"id": "train_043488", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ec442ef52052ed4701470609029f99e5dddb5a23"} +{"id": "train_043489", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c0b3bb456077b3e56a54d57ff101ff3beb6a9455"} +{"id": "train_043490", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "bd831f3872a1e3bbbb4d768c780346840a99aecc"} +{"id": "train_043491", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "5dd267fe5e55d27fbd3eba0d6a94d487226f142c"} +{"id": "train_043492", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4a5f46a4ec1196637831eab0580c7d2a3c0f5790"} +{"id": "train_043493", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "13e992cf00f82940fd39544570baa1706d3935dc"} +{"id": "train_043494", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c5858e84e1088c06ed5ed6ed3e7a9c8552e9da82"} +{"id": "train_043495", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e7015c89be1af488babc02a1446a1e321f73de5c"} +{"id": "train_043496", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6c677604c39e47df42e2c72ec618e3e86fb3f413"} +{"id": "train_043497", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ef5f099553e035b27d8365714faf463e8228b175"} +{"id": "train_043498", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5ca7a2112da029e3f0d3c09cded36c56604d7c75"} +{"id": "train_043499", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "19cc3eff372901d30785eea18b308cec96c47d88"} +{"id": "train_043500", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0c637d7c31f007706dcccf84bc2855a9afe3e18b"} +{"id": "train_043501", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0d327cfa51a5b2c61da65cf162e6e2ed40f5a635"} +{"id": "train_043502", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "cd00b9a27e67455f45f5ba813678807cd1f7b214"} +{"id": "train_043503", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "79046f5e81eaca49b9b4ec0f837d891849bda8df"} +{"id": "train_043504", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "fbdc77a7e1dc080556f730518ca6b9cbbfd1da5c"} +{"id": "train_043505", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "dce167984f4d7f54b19837e0d43bfb9504e9f88d"} +{"id": "train_043506", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9dd6e4246a82398840e1f794d3312d20d9302b9e"} +{"id": "train_043507", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "221cdeab899642ec893b4aef3130c0dff47b6fd0"} +{"id": "train_043508", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "cd6be4723b4d9926eb5db3ee1bb0a75a799cb66e"} +{"id": "train_043509", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "49041da8f311b49dc4654339926b0a597acbc838"} +{"id": "train_043510", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "fedb368be49fbec8c62b34c1b605d94ba2394c90"} +{"id": "train_043511", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "704c9d0f524080900706c70441861c65a7fbf2a6"} +{"id": "train_043512", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "aaba2fbce369d579210c09d888d226e355c4ba23"} +{"id": "train_043513", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b69df4556637e279e86b74df2a736fc7e7c77e09"} +{"id": "train_043514", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "7d56f7a38108bf65398218c8e4a92c376491dafc"} +{"id": "train_043515", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "be0cc32fd871a391708e4f659f893fdb1f69c837"} +{"id": "train_043516", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "af63c90a6297154c04e1c426669783e729531a85"} +{"id": "train_043517", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5c74e3e3abf721fd7308e9f84e4359f937e08ae9"} +{"id": "train_043518", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "4af3701aa3a208bd5e7349876ec93ae203d32c6f"} +{"id": "train_043519", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f41b0a0866fc45b88df4b9cedeb02ce739ecd30f"} +{"id": "train_043520", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "4bd149a2e50eb3f9943c2bb1c4e68146c258be99"} +{"id": "train_043521", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e75a9755a29ef0b0c1f81fdb8bbcb99cf7edb40a"} +{"id": "train_043522", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f7274a9e2a0d6f52a9bf2fece4057a5606625473"} +{"id": "train_043523", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "08203c4312e2c36e9981b6135e45279c75811685"} +{"id": "train_043524", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "c1465ca9dccd88798d53f72e45380ca3b0ecc673"} +{"id": "train_043525", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0abe8ed8ae83f44194a7c06b808049bf280e5188"} +{"id": "train_043526", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3f45624d9e786819793f69ec97e6eae807adbe6b"} +{"id": "train_043527", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c1c0dc043aa38f08e0a5ff4e76604828109d8654"} +{"id": "train_043528", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "00922c0644610d2ef71a152c29572c1695a16a88"} +{"id": "train_043529", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "019c3e76dd719d75b9483b03537cb2e624c67134"} +{"id": "train_043530", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "886aec2d53f980d0954137445c6e6ceec2299c38"} +{"id": "train_043531", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "63d08f372161c1bb94b4af29ac766698349337d3"} +{"id": "train_043532", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "17f6f2235e8ca97ecd55aac6321075fef4117954"} +{"id": "train_043533", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "8f5b54531539cc413084f17485ba8b3f29e9b65e"} +{"id": "train_043534", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "58232ef53ff9b8bb73c23f26b52ff2fc6b7ae3f6"} +{"id": "train_043535", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "26c3382266a7aad38ea354cac6746f491e5503cf"} +{"id": "train_043536", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "54d2f174e43748eae94bd2244cacaac0172e58f7"} +{"id": "train_043537", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f83bfaae89c51211b6cd7889edf734bc86e92715"} +{"id": "train_043538", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "424de93bda33cf514c005fe63f075f89c272d405"} +{"id": "train_043539", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f839f746bc3064b046af2db6f13904e8cf29aa96"} +{"id": "train_043540", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d5fb5c9d208f692cfa60d25feb941ed1eb81a7c7"} +{"id": "train_043541", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7c1ea7650e87265110f7a629f0fc5c9c04c5d7b5"} +{"id": "train_043542", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6a8a8025249692ce45f976ec01ccd35f2a54186e"} +{"id": "train_043543", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "67da61cbe91f3a720777b375af6d3597d9e5b661"} +{"id": "train_043544", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "500e4d9ce3d8edc5c737859ab0892c5371c95ddf"} +{"id": "train_043545", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "89709c1afb1d85db53f088dd5361634adf11f42f"} +{"id": "train_043546", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0d8adcc24b9a37346b340e2a3c176ad6253d97b4"} +{"id": "train_043547", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a6379d8fc7f63622bcb99aaac2b4dab767c5291c"} +{"id": "train_043548", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "46efd3880c1de69a9a85e712d64c4ce2cf23e5df"} +{"id": "train_043549", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8f7f19c7c1d939aab0683addc408ed3d2ee480f5"} +{"id": "train_043550", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5094bea5ca9e9dd08a62275a12a92b24b2bca97b"} +{"id": "train_043551", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f5997184a50b927851701fd99f4936ae21553d6a"} +{"id": "train_043552", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b8468d58a956a5877081f619e128725b2fd35ecd"} +{"id": "train_043553", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b89c3911ca8244619cd016cc0585f9b1742cd7f2"} +{"id": "train_043554", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "015625ac07c7e5029c63ee11ad2d623261f51bbf"} +{"id": "train_043555", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ecc4802494ad4bfc7ec73e2e496936819a63ab55"} +{"id": "train_043556", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "638e49f866b4b6c5bb4afa21331eb5fe8be5fd2e"} +{"id": "train_043557", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2f2969c46e47a114470383daf7245542df60aac1"} +{"id": "train_043558", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ee61c4258ad25365577090930ff14cfad1591ae4"} +{"id": "train_043559", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8d137bd5f20cfc94286b806584d386e033b00d18"} +{"id": "train_043560", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "027eb0d7c4e9f4e171be0ca359c8b211a2338ba7"} +{"id": "train_043561", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d99c092be5d8f78ff99fae2d748a5f98a6a51b6c"} +{"id": "train_043562", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "98fa5353bf2b2de9c3006907a0a1799e2a0b45c7"} +{"id": "train_043563", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "afae33c4d6c4b0e02f16d4cb68a9e82568b389bf"} +{"id": "train_043564", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9d02a757a1734e7079e4aa261e674f1bf0f0ac72"} +{"id": "train_043565", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "38f8337fa510ff8cc76a64be69c5914b7a5bead5"} +{"id": "train_043566", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "68768083e7ee41079a175c9723263f25631d5626"} +{"id": "train_043567", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "39afc514d566f836637ac1ea20540b47f5384109"} +{"id": "train_043568", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "73a5e2729272c16ac46a1783d34eeff40471b337"} +{"id": "train_043569", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bae808f8d86c6557e37ea490299b947f8390f1bc"} +{"id": "train_043570", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "adfcbe2f3e962aa6de4ccb9e4f9e91f07123f8d0"} +{"id": "train_043571", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0b6443f939233373810e5e12fe525269d437e397"} +{"id": "train_043572", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b5ae0d40c35e307879e11e2d3d6c2840ce7cec2f"} +{"id": "train_043573", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6a88082401a612ba4382dd372dd66b825ef11abe"} +{"id": "train_043574", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5de52ddf3194776a11dc0c21238af88edda93100"} +{"id": "train_043575", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3deb6181b533b6eeaa370c9c6c71c8fd068243cf"} +{"id": "train_043576", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c680235903a664e96495f36a169bed14b2cf9e71"} +{"id": "train_043577", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8ec4b3093c1557af91e42e463919b69c52dbbb4d"} +{"id": "train_043578", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "00a039d8f2879e5014d976966c4d424eb1bf02ea"} +{"id": "train_043579", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "361050e77b8aa6e9614e86b51bbd7a5a5ce8ee0e"} +{"id": "train_043580", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8ec4b3093c1557af91e42e463919b69c52dbbb4d"} +{"id": "train_043581", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "727514e17ac581fb2f23e9084030341aea143873"} +{"id": "train_043582", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "fa947277d0603b74a1c127a5598d674882ffe7e7"} +{"id": "train_043583", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "cc29d191909b1333fae5e465fc51b00487dbf7aa"} +{"id": "train_043584", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "28d566a3a925b295e0b27f35c4e3adb8f43b18e1"} +{"id": "train_043585", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "12dbf7db54a22e0e5d123d159f83937cf9af9e01"} +{"id": "train_043586", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3fb57f9f27a01dac34d24e745c6cba72710d3169"} +{"id": "train_043587", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2490dbef629598c6732995e30a1e1f48f2b25bca"} +{"id": "train_043588", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1c29f4adf9e1603b2553af197b56a5f92e894e56"} +{"id": "train_043589", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d31c8265d72550686fa655e561e3864ad2481ad4"} +{"id": "train_043590", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "52fe928a3141a36cf7d285de3c4bffe1c9fe2ba6"} +{"id": "train_043591", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2493c0cc4689eac58c0342f70cb56c2949a54330"} +{"id": "train_043592", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "17ede8f9c8c3a8601e26b9337a79b15bc7f3e102"} +{"id": "train_043593", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "00d0f2772cf9bc7d77051bcb7453ad486ef7edf2"} +{"id": "train_043594", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3c87174a2e8ef2c212989d0e5dfb75e869e3fe97"} +{"id": "train_043595", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "d6e417482b629473fb70ae2216001dd082b79ab3"} +{"id": "train_043596", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "10dbc7481460f3b8be4201c65e537364441c12de"} +{"id": "train_043597", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6c50a9daf8e3f849362d13450a5b87be6cbb59b1"} +{"id": "train_043598", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "02985baab6d3c15f6e9a71e91c28da93f118d4a6"} +{"id": "train_043599", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "24c2ccd5cfe97460623e91e63339f34e1575ecfd"} +{"id": "train_043600", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4cf5a6d31512aa1bb85cb4d473975bdf7f9bc4d6"} +{"id": "train_043601", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9af055d778a6236c0f260faa06ba5a3d18d20a37"} +{"id": "train_043602", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "27d70833adb7704ca92c5585c0d37ec9995d720b"} +{"id": "train_043603", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "fd334eb19fc427920ac50ec68c09fec9a63c59a4"} +{"id": "train_043604", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a9b5bf740d9b40e1ab5aebc447e9b743f83f9220"} +{"id": "train_043605", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1664a0a99e8ddd640fa82ae8c55d16b42ffa5a19"} +{"id": "train_043606", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "71352c68189b6009a60a70fe4032504e0064fbe9"} +{"id": "train_043607", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "941781b3f4f4acc54f23df10241192476848c2bb"} +{"id": "train_043608", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0aa4fc6a6b8590a530258769f20af7d10ee03d3c"} +{"id": "train_043609", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "50b74975d083a71ee7804c9defdf1a75b02957f5"} +{"id": "train_043610", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e8692c5e5f53ca0ae7f8bb2649fd4fdbf08fadd7"} +{"id": "train_043611", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "50f6b4300d1d598922ab97043f14de16a0ae75d9"} +{"id": "train_043612", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "51d05976f455fa7d085a6d91bc5d1f8ffce14408"} +{"id": "train_043613", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a2b04111458c333ad8c1bd8c3921819a93f36495"} +{"id": "train_043614", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a4314b55238003dd4f8784caab5e8d76bba03e86"} +{"id": "train_043615", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8607a99257d3bf04bf7268952f4d7081d5a1f51c"} +{"id": "train_043616", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4b1bd5b7aaecce47e5d377ea8a762975747e053a"} +{"id": "train_043617", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "40bd0598e0bd74034a0a3fabf610d1acd9a0605b"} +{"id": "train_043618", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "32ba8d274825938c0083bc7f44e6cc806b694539"} +{"id": "train_043619", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0a5c738fb38d50433dcaa2bdf36a7633445a88f9"} +{"id": "train_043620", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "2fc1d81b8bdedf0fc9d5cbd0bd5c412d7adff058"} +{"id": "train_043621", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "78239f9c1fc7839798e5fd2a814609e3061e38e1"} +{"id": "train_043622", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "094a8412e26f400f1fa42c2f4960319ed4abaedb"} +{"id": "train_043623", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "368c27bdb40c4577f4b3ca72aec9f297294b6550"} +{"id": "train_043624", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9a914fb6862dedadc31ba4b4470126068c2fab97"} +{"id": "train_043625", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e26af2152d15875eb87731459acae7633120f97a"} +{"id": "train_043626", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d58763abf56a580506f0cd7e7a82de4dad6e29ca"} +{"id": "train_043627", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "41fa997f95f073d97eff1b833f0404fdf5d4834f"} +{"id": "train_043628", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "38dac40732734c769e0e715fa457f43ce70dc1bd"} +{"id": "train_043629", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "98d5f0d96eca6acfc75309cacad790e083da8ed9"} +{"id": "train_043630", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "de88668767f50d6ef1fb4dee2b35e6a2dd51c568"} +{"id": "train_043631", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0456f95d2f4f525a325c072cb47e9c5f03ea977d"} +{"id": "train_043632", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d755225b68b6f34c7041840248500acf7138b939"} +{"id": "train_043633", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "678406b795a40b3133a182365262623cc5424e04"} +{"id": "train_043634", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0559b4150abf6c9396a3fb5e7d15ba06f1f258a6"} +{"id": "train_043635", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e80510af113eccc1dd930803af836df69e1f3038"} +{"id": "train_043636", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "284fe7e3cf950818fbe28dab4bbd58b57f5115d1"} +{"id": "train_043637", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "95733a8785a91f0ddf19303a47b95aafbeb3c59a"} +{"id": "train_043638", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7119422ca5a83286d4ac6e192d1a7138719b42b4"} +{"id": "train_043639", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "22701467f9e4fd979cceef0484a0046e75ac80d5"} +{"id": "train_043640", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f87232cd658c0a3cfd8af758248fa96cd5e5cfce"} +{"id": "train_043641", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "7cc21443b619546dc0fb22f81233ae3e0518b808"} +{"id": "train_043642", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ffd9f35bb4c55dac16ad425bcaf82688232ba969"} +{"id": "train_043643", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5a886033f5c472fa095425e06bd072e5c5a2e031"} +{"id": "train_043644", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a3ecedc758aaf938aa3791d876a2499f4e5dbdb1"} +{"id": "train_043645", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ed39a75d57b1299778c3a633b68f0c911c8bd44c"} +{"id": "train_043646", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "b880b29313295a360f45513db7960d4c157019f6"} +{"id": "train_043647", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "08fd6adb2530ece689e2e3b501aec81239ed10c8"} +{"id": "train_043648", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c40cc7044dca3166976cc10b1de9d992ac38e4aa"} +{"id": "train_043649", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "83b7a60cf7694c6621b25cf6d0c790847a776b2a"} +{"id": "train_043650", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3779a3e3fbc27b2775fad3b2e8168aef5b6b0e4a"} +{"id": "train_043651", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d6cd6059c1b00e346665ea366dd85629cb09904b"} +{"id": "train_043652", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1c681c38def0b361cd0821764dd5886e628a547a"} +{"id": "train_043653", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8085cdf29d09e2a833ad99dbfbde8ffd5e431be9"} +{"id": "train_043654", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0a8a7f385fedfbd233a59e61ee590b82e3389639"} +{"id": "train_043655", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2f60fd62d38e7e9da1770770afbaa4f0f314d1fa"} +{"id": "train_043656", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "05718999ebca307ffece29b0fe3970e7c2eadfdf"} +{"id": "train_043657", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d32d7db05d427279e9b84f1e655b3e85ab798bc8"} +{"id": "train_043658", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2145c76282085b50abce191945854a9997a40340"} +{"id": "train_043659", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6dfaeff1716776bf0aa7caa8035d203845174869"} +{"id": "train_043660", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3cbb2ff0c24ec5426d8922ec13c47b211b4f77a7"} +{"id": "train_043661", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "84346ffa5480c22779899a5e0d2514d1ea4bf885"} +{"id": "train_043662", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8e029ead29076d49c02c337852c065ae2278325d"} +{"id": "train_043663", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "47d5e3924f1613372e402987887d81de1499ef74"} +{"id": "train_043664", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8ee68a018f6d1f097e1f3a9ac271f050395e94c2"} +{"id": "train_043665", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a3c08d729eca2e54584da85ef63be20e38ec563c"} +{"id": "train_043666", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "9e62a7d6d20f77e80485a4bb93d616b6d1fdf437"} +{"id": "train_043667", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ca9cd464027af89775add5f9c6e80ee04c9d7867"} +{"id": "train_043668", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ab822014273c8d3faf52440957702b3786ba3d52"} +{"id": "train_043669", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "317e55377b0007d1ba1ed859923d798ea9303e66"} +{"id": "train_043670", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "eed824d368560de5ac142f0821c269eee221578f"} +{"id": "train_043671", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7523d93b37ece021ca8668105d2e9c11226480ce"} +{"id": "train_043672", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0fa2a993a2436f6e557e0273d001253588f621d5"} +{"id": "train_043673", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "330e2ecc01632e640f85baa2299ec33d1fd9ced4"} +{"id": "train_043674", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "780f2a651740ee57015068fbd46c96f28b5db811"} +{"id": "train_043675", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ecb227d4ee4c85c7e1308ce6aa2693ad097ed129"} +{"id": "train_043676", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3375cfe7c33f1981a8e5796770ce28f706ba5785"} +{"id": "train_043677", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "19ee3ab6ba73475970f15cfb4706ce9d144e2862"} +{"id": "train_043678", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d61010e6c71899dca870be814e7900e3b666a780"} +{"id": "train_043679", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ee04eff9f60b141b71958c80d497b33b9c98cd50"} +{"id": "train_043680", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "aaa059099ca34f470740dc63d5cc35a2faf30a14"} +{"id": "train_043681", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "1e6728392e022127590cb433e26afd3a2d0bdfa2"} +{"id": "train_043682", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8d360de7f18d067ea53feb326a5a62a4988d5fc9"} +{"id": "train_043683", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ecadab1cf90b696fb283bcd65c8532e6f509dbbc"} +{"id": "train_043684", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e259007daaeff209655608019e1727b7ceca2de8"} +{"id": "train_043685", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "07553415fa80a7a82e1053504056d6eb1bea7d05"} +{"id": "train_043686", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9a3fa15cad0888d8b3a180ba31bfb2a3dc22d435"} +{"id": "train_043687", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "5b505e2b78cca8fa4bced8c07fc87f8e78c5260e"} +{"id": "train_043688", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5007962380b1f1c3a2352db97a37b52adbf9dfbf"} +{"id": "train_043689", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "94ffff0149b130eeaf320a107fd848b17060bcc1"} +{"id": "train_043690", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f8622622280a5c62bf9a0cb63df1a0e0b4ebcae3"} +{"id": "train_043691", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cd77521a6ffc8f64be3d729dbc03579fbe11bc04"} +{"id": "train_043692", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2e932b57a5369934d2c93f012fae15fc5dfcb4a0"} +{"id": "train_043693", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "d09b08a2ce53401aacb0ec34bbe38f8b257baf8a"} +{"id": "train_043694", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9df886db76864475bcce700150236b8b057235ba"} +{"id": "train_043695", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a5b0dffb8e298ed0f06132f477246d85ce61410c"} +{"id": "train_043696", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "3bbc2c9ad1c93fb9a05901f34f540733611c8b3e"} +{"id": "train_043697", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "fa048162a5fde15624650a8cc84ec11b9ce610cc"} +{"id": "train_043698", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "fca1020abdc87de21b141ba090853af5e7becd91"} +{"id": "train_043699", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "18cdcd4b48fd64129e42d1d18edf2d8a985ec520"} +{"id": "train_043700", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ff6294c040e7b04a1d62f61fd35bc626bba4f892"} +{"id": "train_043701", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f8a22bb0284f28b866987cffbf036e9b02266ad1"} +{"id": "train_043702", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "893cf2e61f3fe8f5d88d1e7930d6e6d17b173ad1"} +{"id": "train_043703", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c716715859fa4f18eac6dc579f6faf4a6baf735a"} +{"id": "train_043704", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d53ee6e5fff4ed52804da1539e1d83d503ef9d6d"} +{"id": "train_043705", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e4cec25cde99780f525c3c7066aff5076ae08131"} +{"id": "train_043706", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "df1fdc2b58f00d0e8b3527fd0864d244e0878370"} +{"id": "train_043707", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2c9e084d04ed7f9cabc27703dc43b4a6a1cb4fed"} +{"id": "train_043708", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e4fc9fc2edcf22f7c8666ba0fe641935d9eadf98"} +{"id": "train_043709", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "840fc8753486f7854a44c3ceba363d2a39817460"} +{"id": "train_043710", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "11403b3a614ea3413d566708dc8a7e16e4579c95"} +{"id": "train_043711", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "eafa52dd1479dd73d483fcf474393b257cc26dc7"} +{"id": "train_043712", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "68a00b97c7237956d491996506c031a752f3f5c5"} +{"id": "train_043713", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "840e943b126ca894fd9305b78e812ead720074c5"} +{"id": "train_043714", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5aebc4cddf0ef3e4ab831948eb22f2a9af8b6805"} +{"id": "train_043715", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2313d27830bfcdecd3f855be90bff03fa62955d8"} +{"id": "train_043716", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2709160c90c8e21ca59701de193fcfbcb37ffb11"} +{"id": "train_043717", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "6ee686219ba0e3688e1ce09cb475fe696464c618"} +{"id": "train_043718", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "64a5fc91ac3911e01c4ba476a24dd717769cc79f"} +{"id": "train_043719", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c53a66af9e97630ac9576c0e02012077d66d13a3"} +{"id": "train_043720", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "047a409404dc5513b7739c8e10c6dfbb8ad2a578"} +{"id": "train_043721", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "84baadbb2d0352af833948f9a986d4cc696fc68f"} +{"id": "train_043722", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "95692c01b90bdbf0daa65cedd49ad8abde0491f7"} +{"id": "train_043723", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9dd5b0b6af59146721ac358edee1676e04ce6189"} +{"id": "train_043724", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3614b71a194f1862537dc0ccf14fe9125083201f"} +{"id": "train_043725", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ebfbf4735acd59787d24049876b5b5ca8cbffaca"} +{"id": "train_043726", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "86ca7dab7f6f5304c8b0029f27d9d3c57c315e7b"} +{"id": "train_043727", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "65898ebbaa23aca13d2cadccc6463ea489b1c348"} +{"id": "train_043728", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9c18e4083f1e5e9ed8f49bab8f7f6335227fcbab"} +{"id": "train_043729", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b3010c7e79661f1442f378ce42fb51253845edb6"} +{"id": "train_043730", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "23af050a48196d3c839cbfbc1ecf0be709190d6d"} +{"id": "train_043731", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "82787b20d47a874d7f27caab355785e7c87b0d65"} +{"id": "train_043732", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "748cc9644ef5615396c4b9800198ef75f7cf2c06"} +{"id": "train_043733", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "df5ec76d3f76ab131f738166d9b81d9efa7cc478"} +{"id": "train_043734", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2f45e15c2f3a364ac95395aa35299e291de7cdde"} +{"id": "train_043735", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "913a576db1c3f3a3fc3e19259d2cbf7e3b0105ff"} +{"id": "train_043736", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "badbaed4cbbd204ddc134abc2a6d3f285d1ae5bf"} +{"id": "train_043737", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4e46300eac24ab803b0cc6a632d2af4fd08e0be5"} +{"id": "train_043738", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "20b82f5e4b6807ee6e076875a423d38976845ece"} +{"id": "train_043739", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "022e09f3187f086f19fda688b0289d765174f25f"} +{"id": "train_043740", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0eb4f285c1cfa89b84d7529cf282108c45a8696e"} +{"id": "train_043741", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "db583044c0d5b87a49c1c64347eee8184ce724f3"} +{"id": "train_043742", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "187c367208e489acffbc283f13ba94b6b050efad"} +{"id": "train_043743", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "2b5fce926d1d2978eeba691174fd42d3a40d4ad0"} +{"id": "train_043744", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a15fab98577b6be1c31718426b9d890c9bff817e"} +{"id": "train_043745", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c2745fa2094a76079d86f80637fa50622731fdc8"} +{"id": "train_043746", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9efe89c01efc062c4f9da16bde9063ebd868ecbf"} +{"id": "train_043747", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4b209c59cef8a3fcfc5097c9ba887d88fdc95073"} +{"id": "train_043748", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "19bff1069dd84304089908959a1c2195de1b25e6"} +{"id": "train_043749", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "898c166ef593033285efb1f9ad1994d249d12fab"} +{"id": "train_043750", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2059e5c1317333502c960a79117e7ca8a46308aa"} +{"id": "train_043751", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "608055982bf400f0a2a71ebd49e8313255606acc"} +{"id": "train_043752", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6c4e353e6f8e469c552dd2190d9d710b44a3ab9c"} +{"id": "train_043753", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ef1277b93b77849cb02897a44a7e60c09e8719bf"} +{"id": "train_043754", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "28ddbb13eabfc5f222f6c6c1548deedeb592a2d7"} +{"id": "train_043755", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a6e5d5c2d29e3c918eda0d329ed1ada007660206"} +{"id": "train_043756", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "bdbae3857f1a3acea37413d7c5e758caebe855e9"} +{"id": "train_043757", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "e927d7a23de86170a7b23b38114e04ab50d8a988"} +{"id": "train_043758", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d45db4bc6395e3668f4ec585636d829fc184112d"} +{"id": "train_043759", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0899d484a49a1d52fcd139ff2050981c0d475c62"} +{"id": "train_043760", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "306694270d66ee3c1bd9cd7993d1ffe331f93626"} +{"id": "train_043761", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "67e484ec47b691347d0519615f83306daf855ba3"} +{"id": "train_043762", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "89ddc9f208f968a0f2ff05f333d94dcb9b9527bc"} +{"id": "train_043763", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "790cc37eff4e1ba4c62137784683fdcc35a711f7"} +{"id": "train_043764", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e7ca2acb3ba0079feb35eb7b6d6fa697fa134cd7"} +{"id": "train_043765", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "91475e0c037933345af7eb10756ec80fa579378c"} +{"id": "train_043766", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "b4001cf60cf87dd56978684485edd273c660e240"} +{"id": "train_043767", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "efb091fee4d570161f890fd442511729ab50135d"} +{"id": "train_043768", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "695b71e52cd13da996557dcf72f967e7eb874252"} +{"id": "train_043769", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b86a227932de3a28ca5b1149cde1f524d5ff51e7"} +{"id": "train_043770", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ec77b5a3f064de960ecdb887f54e967162af1072"} +{"id": "train_043771", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "10e861a15e03b5051d4f7e16f38c89b43f27081b"} +{"id": "train_043772", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f2d380f8dcda5c6a676b5ba358db4c2301b94b9c"} +{"id": "train_043773", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "32c7ac9b07a94afdbe4de7eb3596da0dd7de7c8f"} +{"id": "train_043774", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "25d6353adc35801f4992e43bb3a35ff10fa51922"} +{"id": "train_043775", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c89ca8d94b2cd549e981a92d85f386f903af4f17"} +{"id": "train_043776", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8e1f48cbd64f4bbf14e7dc1e7b75185acd281064"} +{"id": "train_043777", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0edf02213cb2f8e6382ef7ca720ee722a094c968"} +{"id": "train_043778", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "64639bd9d60c85378330418071d5291722392384"} +{"id": "train_043779", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "a57ec6cca6327a651182a4c967d6adabcbdfa841"} +{"id": "train_043780", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a2ff5d8cd646db9b690724acf0914a35aa5275c1"} +{"id": "train_043781", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7264d42e56becd04ef0d270a2ef8754720ea8ca8"} +{"id": "train_043782", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fc6b9dd11fa7ccf7ced6b508984dd37dadf0e604"} +{"id": "train_043783", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "3165acd20c89412874e9756274df5334c5b330eb"} +{"id": "train_043784", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2a7dc411f37f3cabe98e65c1dded8c221c3d4f22"} +{"id": "train_043785", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e9cd2cc7d4994b3d522f59cd5cdeddd2d8be4644"} +{"id": "train_043786", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a7d8f0a054c1da11daa0a88efd396278ed750964"} +{"id": "train_043787", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "85c4c73491d08f834adf309515375e203e83b3fa"} +{"id": "train_043788", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "42e83934894b07d2d61d2011c810b872ae43b620"} +{"id": "train_043789", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d435f7183884d2c2b19a914fe372e1d6e1c042ca"} +{"id": "train_043790", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0e0fc233d72a31b6424f1fbf8b89bdb6b641577e"} +{"id": "train_043791", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d0c9dc7ee15bff60d6b822501aa34d0f6b789ab1"} +{"id": "train_043792", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2a7ba893b672cf8655d8976c1a626c7cdcf87e65"} +{"id": "train_043793", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ea1abf059820a2193fb6059fe501cd43a71f659d"} +{"id": "train_043794", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "fd7fe7d29bb1c1d98ba5cdb01cb3a2f10787d5b1"} +{"id": "train_043795", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "06c978aa123b25fb86d7ea6d580364408938312c"} +{"id": "train_043796", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "268340d7d58cc719ac7a25118690b4f652248789"} +{"id": "train_043797", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e95c9de8c6b6d3c44bf668201e353b3e51122d1d"} +{"id": "train_043798", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fda43ba2aef596e19ac241132fd3ac1e33e6da49"} +{"id": "train_043799", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f30cdbf1278246c6a063a43937f8328ef28abf0c"} +{"id": "train_043800", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "58c816fa1c972ab11fd1a6e924a234b84fdf9483"} +{"id": "train_043801", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "5db4389c43f555fe825261ca79d0cc5a2ae32676"} +{"id": "train_043802", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b2d79f1601d08ba455c0d76d285a90c904e8ae4e"} +{"id": "train_043803", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c6348fdddade64bf80b9a81a58c25911d328912a"} +{"id": "train_043804", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "16e5d790654ffde3832aa95738f605281b8e3ee4"} +{"id": "train_043805", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0e41438e2fdc8cce8dff5bad1f66a0c7b8aa6056"} +{"id": "train_043806", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "80853098eef4c055e13924d08f1f07fb32d47872"} +{"id": "train_043807", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ba3532ba3b9015933b00ab517554cf21d6436398"} +{"id": "train_043808", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cdc367a5dbcd938e715efdacab8981dae64bd9f5"} +{"id": "train_043809", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6e5ec996bba70410cb920f80d144490529b09489"} +{"id": "train_043810", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "016ee965bf128ab58386bd98e14d6761f58de5e7"} +{"id": "train_043811", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "eef57603c76360572a33f239328d13d3fae5b361"} +{"id": "train_043812", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6b39e468638ec98504a51b082399fcc1910c26fd"} +{"id": "train_043813", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bd56204533f53dd200aa1aac29854a5042259327"} +{"id": "train_043814", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7eaae5d83e7f89be478b5dec96c1555783de65b4"} +{"id": "train_043815", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ad0c2f1a0aa0f137778a47d9c395087c1f48d7a1"} +{"id": "train_043816", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e8c59c7e5e824de0d9ea6255f971776f9e8c3c0c"} +{"id": "train_043817", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3e9472a5e08651c923acc91cd7585b80c63dd116"} +{"id": "train_043818", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "23098bf652c2e843ba81484e2223249de3ce1249"} +{"id": "train_043819", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b9c61326fe9be13e1a9ba2ec760bd1ec4b216dbc"} +{"id": "train_043820", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4561caef5acf8d505b9872163d3ce322b60a635f"} +{"id": "train_043821", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "fe408ab6b3d1a97730dd41cb26aaf4848f47c505"} +{"id": "train_043822", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0148fca5e5862d7b243a161d211d1b0c4f2969ee"} +{"id": "train_043823", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2a0124849579e8c28fc06691e3bc49ae87b4d85e"} +{"id": "train_043824", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "df549922e062460e57d49e5e316be49a24609f15"} +{"id": "train_043825", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "08c96256f76a78aff7deb8de3c037a9ca3d58070"} +{"id": "train_043826", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "543a05034bc5944ae9ea0bdff4493487664afba5"} +{"id": "train_043827", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8bfd8b327497b413be18411898dc965dbb99d432"} +{"id": "train_043828", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e58b4795640a7938203895d5ed5250d6add4e866"} +{"id": "train_043829", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "02a56219a402b28f1540a0b0d22295aeef09a2fe"} +{"id": "train_043830", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2b86de57dc8983f279e9cea677bb888d1746ba55"} +{"id": "train_043831", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3d093defc1a6fab80aaf42db1ed9b1ca9dd23ef4"} +{"id": "train_043832", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0c54fb358d2d838953e8243ab38acac79442e5eb"} +{"id": "train_043833", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2147cfaca21582587128846374359883238998ea"} +{"id": "train_043834", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "22199454e2363ac3e79ec5e5ec38bcfdb9d7d57a"} +{"id": "train_043835", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "cefc8144dced5f7aad1fcd9a7189ddb856066266"} +{"id": "train_043836", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e32cb6c9f3d3705863329449ceb2a27756085e61"} +{"id": "train_043837", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "792b6fe90686b7d0ed4fba7527ec8d2d2541998f"} +{"id": "train_043838", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ab25c2300ae7e1cf68df21c1c28e718ac00754c8"} +{"id": "train_043839", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7dcd24449e0d9bab9eba01434257b09506ba419e"} +{"id": "train_043840", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "0575f85fe38f726e17d7bf3d932a39eefaa8a96f"} +{"id": "train_043841", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "32c8d5041e375374749d65dc5c1838842bf14a6d"} +{"id": "train_043842", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c304778b8614498cdcef550419cb9af80b6502cc"} +{"id": "train_043843", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "893498c0a1623170d740b5a63c4726bdb47b22d4"} +{"id": "train_043844", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "06459876234ce5458f3aa1cc9d1eda4b7e120161"} +{"id": "train_043845", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "38e24cb070c541edae37067db6d0b9c21c514839"} +{"id": "train_043846", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "aa12c88a2d5535d392f5bb246d5d7ed1a0f092df"} +{"id": "train_043847", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4293c3d0aea19fcec2f65d68a193c158a32c8890"} +{"id": "train_043848", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2198d703ca38e404daa9517c4139491cc32a073b"} +{"id": "train_043849", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "38a0a897c1b5cca0297c87428d2df1cef5949a13"} +{"id": "train_043850", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "919c071cdeed95e76ece9f307889aa6a9996324b"} +{"id": "train_043851", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7ffa864f855b58aea74abf018519244f460fb506"} +{"id": "train_043852", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fafc5979dd20d5714b72e850f4cecfb0140abd07"} +{"id": "train_043853", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3fcd2f08e097fd36735bb9ff6ca676331eb4e135"} +{"id": "train_043854", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7add556a8c542a476fbba624f38ef4dae5443746"} +{"id": "train_043855", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2a530e3f601dc760a376e34f857211c3257f3332"} +{"id": "train_043856", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "b553fb1a52e129f798027475ed1bb19d00fdfa4f"} +{"id": "train_043857", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "6690318fa98d390f90a49d55b8bc418be55233e2"} +{"id": "train_043858", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "760bb8ddec1709e5b283b11f159d2282215442a5"} +{"id": "train_043859", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f30117639fea108fb12533326dd8a8ec6be92215"} +{"id": "train_043860", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ac6d2df5a8999810028b69133d6425da68a9cdaa"} +{"id": "train_043861", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "aaaa28bfcb6343a7f820b6de26f8c408c517739d"} +{"id": "train_043862", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "53804991b1642514958faed52d7d0429fab1cb4a"} +{"id": "train_043863", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f1586beb7692dce6c81751bf3353a53627fa3387"} +{"id": "train_043864", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "755e88a773c9d45f8a12df79d03463083cff1be0"} +{"id": "train_043865", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "c2eb1ef0389131ad77684588804ad02ca6ff374d"} +{"id": "train_043866", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "616cdf203fcf7883fd62dc3cea1ca7a9c7936194"} +{"id": "train_043867", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "890018430f7a18fc8a84cbb72ced9f4ca2a8f612"} +{"id": "train_043868", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b1127ed4b3db82d9c948566622b08fb398580119"} +{"id": "train_043869", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "35d11af3d4d896444a088822c040eca4d3bf5b67"} +{"id": "train_043870", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dd50c5245af94ba7a2d6e0ea62ae58137d1bc12a"} +{"id": "train_043871", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a76fc91303d9f809da1485c509d71bf70c00fa74"} +{"id": "train_043872", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0d39da68fa1650c5b422c165830deefd92c04bfd"} +{"id": "train_043873", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a43586314b27251f7cce36b4b7c9b918bbaed1d5"} +{"id": "train_043874", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "819e4932fca50f69c8bfa3cb0d03438cfb45fc07"} +{"id": "train_043875", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1f512a98a49d700e942b31ad64427344333b9704"} +{"id": "train_043876", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4dc4800412318dff6bc3ac61233e4d79a8c0e490"} +{"id": "train_043877", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "08b41d52d88614891dd26ef46877ca4a8f02a69d"} +{"id": "train_043878", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2ea2075bcfa1eb25860325f43761feb5bb135766"} +{"id": "train_043879", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "148def64fa7ca4d45bc2f6e87e3d6c41354ea0dc"} +{"id": "train_043880", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d85a9369eeb988c8e7624334f70cfbaa362697a7"} +{"id": "train_043881", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "79a902c687acb20edcd27bbf82e972cb232d460e"} +{"id": "train_043882", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "991e063358904cd6075fdf7c8e6702051114e087"} +{"id": "train_043883", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "bdd72856597bdab9f988d7172e209fbfbc6da6ab"} +{"id": "train_043884", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "86df490cab05c156e863b16ff05a16a70231557e"} +{"id": "train_043885", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f276ddb73191166e373ce24186a45b2576f6655f"} +{"id": "train_043886", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0cc615f989e17d797951782cab4d936e926fe873"} +{"id": "train_043887", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "33df3fde93e3e776e9adaf31d77e0546442971a6"} +{"id": "train_043888", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8243fb2661d862156e3d1550d19c5ed00c507996"} +{"id": "train_043889", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "14fbaecd2d21565b5551871163a646ea53ab2686"} +{"id": "train_043890", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "18b0f97dbf1f9df8ac13c3833a5cdff5aecd456c"} +{"id": "train_043891", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c8e69f511f3840c5ba35ffcc1039bd20a12895b1"} +{"id": "train_043892", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "fc74af4c92d2f248306a2ae1b951fe2104ac1498"} +{"id": "train_043893", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9e59f20082931f76b936e49794f1182ddb66c1ca"} +{"id": "train_043894", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f087d8d109c7398c4cfea06a091a90362a7e67a1"} +{"id": "train_043895", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1bba0756a3d4bd69e88a70d01f11eaa61ae8f274"} +{"id": "train_043896", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e2cf6319b6f34e152774c9a5221d8df2108bc638"} +{"id": "train_043897", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "db1ec076cf04f93f54cf303f3b0bdbb0d5e8360d"} +{"id": "train_043898", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7d9a5d59f447095d3cc7af8857ec6a944e2b85e6"} +{"id": "train_043899", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "60b40ffd923f717c986e4543fef069e83ad7a8ca"} +{"id": "train_043900", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "57e96f4e0fde7d16f6b2cf61f6cb62110e5ed999"} +{"id": "train_043901", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c626434ab965d054f6f9cff7827ef026de840252"} +{"id": "train_043902", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "285521aa2ba0c95baf583fa7a7683e6810f510fd"} +{"id": "train_043903", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2b6d4d820d32022f1f9d26751c4e8e1c6ba32294"} +{"id": "train_043904", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fd463cd19762653d1cbf564fa462118e02398407"} +{"id": "train_043905", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "10c83d3941bf3f152da3ef0da8cc35f0e9c17501"} +{"id": "train_043906", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "631d5965abb8d840258fc5b03b07bd3558eb2f01"} +{"id": "train_043907", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c4639f0eab90e303cd908f95b975a8f1744f0593"} +{"id": "train_043908", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e8521ce42704a85cc4efda8ecd66e3060c9640d1"} +{"id": "train_043909", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "cf9dccfa2eca17bfd6ab1ecd46aa226ec2d97c41"} +{"id": "train_043910", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0d0c14b0fb5d19e4b7f1a5bf5b115c71315df3c1"} +{"id": "train_043911", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "c96e0a3e44932fcc1048d213911a8ec2dfe9ab3f"} +{"id": "train_043912", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "58b505483d587aa23e4fcb6635d957215c6bd9e3"} +{"id": "train_043913", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0c643402afa1e78966b6a673a5c2e3c77a561142"} +{"id": "train_043914", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2978319f4ec6109adc3960783f5ec5767176f53d"} +{"id": "train_043915", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "5274e7b08ed941e6457aacfdf8e81c7645fbfdb0"} +{"id": "train_043916", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "71bf94c032e17007d827d2087864f24e8a12062a"} +{"id": "train_043917", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "60e69a0c9fd1b6204cb15339766ca864fc0f536b"} +{"id": "train_043918", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b2fbf978e8c01760f6dd0946bf1cbd7a5be9bb2c"} +{"id": "train_043919", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8cc7210db1202d7036b670ae795f55a9993665c5"} +{"id": "train_043920", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d8c656038ff0fa1cbcd057f7bf99cfe60e0f4057"} +{"id": "train_043921", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "77aaf47b35fff9adf3321ed8625bb8b7c42428ab"} +{"id": "train_043922", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "86809afb6028582588dd949b1e6e3f49bbdab5af"} +{"id": "train_043923", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8a7af8253fa1c1e2b7fc38affa7037f70498d4dc"} +{"id": "train_043924", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6719b528e5df6966829edc5624295ea0fc9435a1"} +{"id": "train_043925", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2382834d250a8351142985d783e7a5695d6290c3"} +{"id": "train_043926", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2c9e084d04ed7f9cabc27703dc43b4a6a1cb4fed"} +{"id": "train_043927", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "df4034ef59a61ad60ebc69c768da1e246b429215"} +{"id": "train_043928", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "3d3ff2e39842660ff24df67bba55ab17890b05c0"} +{"id": "train_043929", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "befeaa61690678fe1f5467e8802a47f68e91ae67"} +{"id": "train_043930", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e3d723e8707a381edd462b7200cf2930aa8c86ff"} +{"id": "train_043931", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "54ceb4bd78d75ffbdcdb9a97af175f538ad3951a"} +{"id": "train_043932", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "dd3c2a0fe0b35920293242533c970a9165decd2b"} +{"id": "train_043933", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f0e343506c09a8b97d6abd300c867a901dd77d91"} +{"id": "train_043934", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1b231b768e9d446c0f856630c46b510d6c746013"} +{"id": "train_043935", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "166dee65d20d951dc516501d45b649880fab9637"} +{"id": "train_043936", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "431281a1ede60d5777b71e7704ff66f856640ff2"} +{"id": "train_043937", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7bfe8373f1ddb7ab5c236a0f94343244af846005"} +{"id": "train_043938", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8ac850ad9580ed1bbe09c4dd9525a53cea9547c2"} +{"id": "train_043939", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f063217e799d1d2b84c270354ee4f567b57fd5cf"} +{"id": "train_043940", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "afae33c4d6c4b0e02f16d4cb68a9e82568b389bf"} +{"id": "train_043941", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c836e8b15d005320d6ee690195f4124d165c8c66"} +{"id": "train_043942", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "77021bcf0f8e45c76b07427ba61286073ba5bf9d"} +{"id": "train_043943", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "98e1d30bf96ce212d8466016c71d163f0653a92e"} +{"id": "train_043944", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "66322ad4f646a58fde2a86bceb04bedaaca2e223"} +{"id": "train_043945", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c275910d8dc22adf762ab950cf37ed1d90cd6a3a"} +{"id": "train_043946", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2ae6a3d2e9b285236b58591d45c310e9acb54879"} +{"id": "train_043947", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "da187dc434fad34d99ea81122701cbaeabb2e175"} +{"id": "train_043948", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d9f3cea41eba3e92a9cc3f406391828e8ee57051"} +{"id": "train_043949", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c337f3a45d6983662ae8b49b747daa10b727f223"} +{"id": "train_043950", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5c553f6ca6d99f422da33934f9ed690572a90062"} +{"id": "train_043951", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "314af58bfc72e0e29fad60a0f3917839e05731c3"} +{"id": "train_043952", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "54aa271b5814297f72c149199630f218cbe479f8"} +{"id": "train_043953", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9f2ac2dd49c9b6f16c129aff9d0fde4fcf888c2e"} +{"id": "train_043954", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a4402b27a3e7360ddcb1a3432cae5799b6e9ed6a"} +{"id": "train_043955", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a99062cc419d9b5a0c44f59914cfe70eded945a6"} +{"id": "train_043956", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ded7e22e5a678409f69307a5df3eba25e9279007"} +{"id": "train_043957", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "64264f4d3dddb49a4ab71d62305a831325df3031"} +{"id": "train_043958", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "260076bd3c915a6439e17040820dc5a3445a530e"} +{"id": "train_043959", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "cb240f13c257c7f8eeae0f7057a4a2fa9912ac33"} +{"id": "train_043960", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "290da7e2caadd9c61617c64348a3d0b280715e32"} +{"id": "train_043961", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ff16ee4a422c26b930ca1947de3229e18be4e9f4"} +{"id": "train_043962", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9716d77b61edc2eee061936f0cadb22271d462da"} +{"id": "train_043963", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e125d5555201cc983fb999491e4609279c2edcab"} +{"id": "train_043964", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "173351ec0b8ca3cea781fdddce09fb73aba7a160"} +{"id": "train_043965", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5c6c5393934e203672e50b7046a528a3d8b6239d"} +{"id": "train_043966", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3d79b19abcce8441c1ce427b83e7fece379a0cd9"} +{"id": "train_043967", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d1d44ec39684f68471b1f427f17d6934cb75babd"} +{"id": "train_043968", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "302ce8822af15052766544a1983cf027d68db59b"} +{"id": "train_043969", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b89df570eb320fe17bc1aae04f89057168ff8654"} +{"id": "train_043970", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6f859af600f6b98f404014c4691e274445f59c3d"} +{"id": "train_043971", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "32a09c7fd423404d7dcb487692b93c1d429cc7c1"} +{"id": "train_043972", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "14fbaecd2d21565b5551871163a646ea53ab2686"} +{"id": "train_043973", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ecf68be4913f3de8b49f6add502a2e3dc4802d64"} +{"id": "train_043974", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a8ceac23746c9cbcd6b8133a880b64561ae8479f"} +{"id": "train_043975", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "5e70966b08f0f1f65ee946bf216b91862583ba64"} +{"id": "train_043976", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0a4d1c125c9fd25f72cd7d5db32a91cb4b84a895"} +{"id": "train_043977", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a8304293f0689e52e8ef15efb1754124be992c17"} +{"id": "train_043978", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7fa33ec53490586bb19afaa2911eae3b3b1ae886"} +{"id": "train_043979", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "948619ad564ab6768035765b37cbd0930f4ed269"} +{"id": "train_043980", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6561337798ec6674153a56b8493b29e2f9a65e69"} +{"id": "train_043981", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9ab44c5fa349f90016d6ea4db997c5491a8b2fec"} +{"id": "train_043982", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d369183bd921d964406216d732765eb96d1ad85a"} +{"id": "train_043983", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "62d4e874ea9c719204233092e949cac36c9e4a2e"} +{"id": "train_043984", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "a7368ae4285fe63defe3ae9604c656af42e2641e"} +{"id": "train_043985", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "bbdabc13985c6b3699a2927142a97d27bad82b89"} +{"id": "train_043986", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "129859bdd348181ffae0a8b190c368b18e86d828"} +{"id": "train_043987", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c90d2ed748d362b96160cc2eca62a3e7c1e24782"} +{"id": "train_043988", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3bbc2c9ad1c93fb9a05901f34f540733611c8b3e"} +{"id": "train_043989", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "4c9897e16869e8936d548258cce49be9df65ed5d"} +{"id": "train_043990", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3775dbf9d8880a1492dce4ef57bf5cfedc9e3c4b"} +{"id": "train_043991", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "e78650cf94093b5773041be25e4bcf65a52731a3"} +{"id": "train_043992", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "05a3b60b83659414730ea69b4c85bcdf6e07c321"} +{"id": "train_043993", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "00f350156cb1c8508984e870a62a24d1363b90eb"} +{"id": "train_043994", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "62ce6fa891fde29b766ff4b3eada804d2a629df7"} +{"id": "train_043995", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c20af2f3a52186bb2d0f7271424489318e2dd4df"} +{"id": "train_043996", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d0f94eccf9bbe0a2814d91c40fe914eb668ff43f"} +{"id": "train_043997", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9a977f0be8c72a0af91f9e71d0ceadf59fa7edf8"} +{"id": "train_043998", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "994440c933f68eeb9ce82ebd77c4dcc8053f4bc2"} +{"id": "train_043999", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "37f971bfd486f0802b9595fca7bd105979c4c66e"} +{"id": "train_044000", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6cc7919f1175604bfebfbe845a0a15a0d292f2b4"} +{"id": "train_044001", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "cadb0fd8dae0e183c5ce9b75a6f1f5e9a2e2e4a3"} +{"id": "train_044002", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "7459607ef0c53e3187e39d075b05249f920b161c"} +{"id": "train_044003", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "03c3e363a603106891367ff190658de95fd1ab81"} +{"id": "train_044004", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "70feea2450d77b079b359de2b8dfc74c433e89b9"} +{"id": "train_044005", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6b033b809ad11ea0fd288ccc28ce03d3d255fb19"} +{"id": "train_044006", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4325d20b7ec95386f5bdc0a6dee4f7b4f1439dcc"} +{"id": "train_044007", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "015748858be2e2328336c4504bd9588e43a9cdd9"} +{"id": "train_044008", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7c47594b69982e70516ccae274b887d3bc5c6c19"} +{"id": "train_044009", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "af9fb3c897591526de2fba8eedfd64dcf75479e4"} +{"id": "train_044010", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f25916642a36bd403f900e52fe579700d4328352"} +{"id": "train_044011", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "64852e7756e467c8e70c6e11e053c0bfbc37fa38"} +{"id": "train_044012", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9d71bb0a6c26cfb83af2f1c5f4dd9d9ab1506adb"} +{"id": "train_044013", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "5193fd0711b69a1036604066bb04abce80a7afbb"} +{"id": "train_044014", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f0190ac75cd51393968cdac6fad154e9f5b98a84"} +{"id": "train_044015", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f22f06fad751d7952d8498a6601fd4434bcef51e"} +{"id": "train_044016", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "3ce3630f153b1eba6b7d9844314b1f4b1a4c7484"} +{"id": "train_044017", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5e03356b68cbb94eb2c5c9d6bb87ea4fe081e802"} +{"id": "train_044018", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "188a396f3d4be5e08d7b4b68263ab139bf6068e6"} +{"id": "train_044019", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "397de0a0711d4e9569a151bfad4cc6e32b383f88"} +{"id": "train_044020", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "cf30cded3976aa2855c7fa0ae34dfee1e77d32f4"} +{"id": "train_044021", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a81bca81c68181875913c69012427f0a12e37824"} +{"id": "train_044022", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "47589ea8eb329f5bc732eb7bfbaae0dce7fe7570"} +{"id": "train_044023", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6e91c974b0ab75ded584bef5c1dac9dbcd867bcc"} +{"id": "train_044024", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0bf2373ab01ce8c36a05f6c1b08f6c9683e95665"} +{"id": "train_044025", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "7fab12f6dcbdc4c84f9e95b03a5c6564973df056"} +{"id": "train_044026", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b89c3911ca8244619cd016cc0585f9b1742cd7f2"} +{"id": "train_044027", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "d911644f5641877f59d5a5c1393fbb51a519d73d"} +{"id": "train_044028", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4bd149a2e50eb3f9943c2bb1c4e68146c258be99"} +{"id": "train_044029", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c9cc07ab9f264c1463316c1650cb4b0d0742ebeb"} +{"id": "train_044030", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "06fae7d915dca9e509524aac9eef01e482ddee09"} +{"id": "train_044031", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3c51dde93e5833c2ed18f4a2708dfd6046cfceac"} +{"id": "train_044032", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "54ffc263254b411531806bb75fe5af89881cb11d"} +{"id": "train_044033", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "72cae0ef2b4a4ebf0d89a75a2f294129d0d477b8"} +{"id": "train_044034", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2957d0db4bfa5861c5560af8bd341a507776f002"} +{"id": "train_044035", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1b498197ad22aa078d377605941b6e491b4502bd"} +{"id": "train_044036", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "76cdfdde061cd5433574ae029ab71c3b811d7707"} +{"id": "train_044037", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8b27e411e600ad770b85fac9c79f80667b78f16f"} +{"id": "train_044038", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "9eca8a806548a838b75133a4898a28014718c96a"} +{"id": "train_044039", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "06fae7d915dca9e509524aac9eef01e482ddee09"} +{"id": "train_044040", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a4a6d11691603b4c1b88619984af7a8427b98a56"} +{"id": "train_044041", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "bff8f18ea1f5562fabf55ea7f7462ce07f907776"} +{"id": "train_044042", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dbbebe903960a61bc681944b10e8d686c7ae5078"} +{"id": "train_044043", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d19b11a64bb699073f77dafb6535e79ee37af274"} +{"id": "train_044044", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6b0d4829439a7303d2a1d0756386388c656dd16c"} +{"id": "train_044045", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4d739aa562ce44b97ff84fa35893c990f79397dd"} +{"id": "train_044046", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7ec73ce22d1cac7b08fac66b2689bedb21c3c2eb"} +{"id": "train_044047", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b8e0c259ffa6dae9076806ed35691c3435fa2b7e"} +{"id": "train_044048", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0c7f92a05107c265b55db4e282c8f35e993f248e"} +{"id": "train_044049", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "14e0c89382f8b94ac4e3af5b4f12da8d6d6923d9"} +{"id": "train_044050", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0abab8c42415f190ade700fa5b1d80153d9a5015"} +{"id": "train_044051", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "17ede8f9c8c3a8601e26b9337a79b15bc7f3e102"} +{"id": "train_044052", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4a6430cd8f6821eece3b04e6fa4fdb95bc692c60"} +{"id": "train_044053", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5bd7730ba45f7aa3e010d1551b5cf94d934049a8"} +{"id": "train_044054", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ca712ade3d3ef418ddfeaf398699cb9a4571c358"} +{"id": "train_044055", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d40a5d98e042c1cf83e2c8e2fbdc92d9fa46cf31"} +{"id": "train_044056", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "6cce94668f3f644857ea18bac2cc9b1796159685"} +{"id": "train_044057", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ab44433d87f3f2bbdedf48b7657f0a747c5a3938"} +{"id": "train_044058", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "6925d6167c8f805aa13bc9ce358f4436562d42ba"} +{"id": "train_044059", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "59245ce78be830f009995abb768a5232df7fd53a"} +{"id": "train_044060", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9bb1bd15886d66d77ed8e1bc0ea087da664e26b2"} +{"id": "train_044061", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "45d5675b5ff465f8dcddf4f8c092d0e8e89c6ae0"} +{"id": "train_044062", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ffe2137e3fa0a78adc1df8423379946d045328a4"} +{"id": "train_044063", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ae6d4a8fab7023c0fce0a97f8471f7ff6c3a24c6"} +{"id": "train_044064", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f66d204619127df84652532250c0e1d90f60d784"} +{"id": "train_044065", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f5d0fd588dcfd2bcab65e4b8c474e419d21d0242"} +{"id": "train_044066", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6f6b84b5285df474e47d01a1b66b623ebf43ec05"} +{"id": "train_044067", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c87c3c591b623eeef946540f973892d5bc16f3da"} +{"id": "train_044068", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8bfd6f5f380c379a0a48cf785d09b666b28500a7"} +{"id": "train_044069", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "630ad727f77dad81b325298228570843025d23e9"} +{"id": "train_044070", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3339f155ec240b898086c448721d0e31d74bbdbc"} +{"id": "train_044071", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f9f8909c628b6fdec2fb3ea155bbba8b32e52808"} +{"id": "train_044072", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e9371d78a44e8e671c34ba862483b26b3a6e7a1a"} +{"id": "train_044073", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0da11b72f051a1906a1a67597f409db8fa042a61"} +{"id": "train_044074", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "756736afc6bccf5a5fb02a495b70469992393135"} +{"id": "train_044075", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8655c7f381305370f444320e5a92ec83a4ace1ca"} +{"id": "train_044076", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f47b7066dd6172c2d7acda12157021e65fa3664d"} +{"id": "train_044077", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "682b140ad8b5b0ef9b40a2c39ece4577d124bf50"} +{"id": "train_044078", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "67d1264110dc921dc18d0440ea96609c0a799acd"} +{"id": "train_044079", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "3f3cac97fe4cdf76be511462d2635bcbfcb6462b"} +{"id": "train_044080", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2c0c1dd586e30a526d1792dc6029576ce47027b0"} +{"id": "train_044081", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "86cb68e54940678c9072aabcbb125a69ab60b822"} +{"id": "train_044082", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "3dc19e407e69f8ebddaf849a86c8f2fc1e59b2f0"} +{"id": "train_044083", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "725f161412cf4a61d030fdbcab438f916c84d227"} +{"id": "train_044084", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "4f327429bfe8b46dfaf74f2de632424f9aaef496"} +{"id": "train_044085", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["agent", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9b41bdd138842e530d443fd6c8b44fa73c526541"} +{"id": "train_044086", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a6fc27abe0245b9a04bfce19ef42c801e26c3a05"} +{"id": "train_044087", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a6e4b6e9ca2a53a0e7cb1b4734347e10b13652e2"} +{"id": "train_044088", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9f1cf54555a5fd47f2be9b1df50d8d9c4207dcdc"} +{"id": "train_044089", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "47589ea8eb329f5bc732eb7bfbaae0dce7fe7570"} +{"id": "train_044090", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "23b8cd57d7e38f1140a2395ce0da2e58ee9261de"} +{"id": "train_044091", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3b5f69c6490a229d8e4170996a086d90fd71c38c"} +{"id": "train_044092", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7a56e9619dc2b5b0749b8627983e35f2c9b26d89"} +{"id": "train_044093", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "d31c31e60ba2bafc5016ae81f976c6a37599a360"} +{"id": "train_044094", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4983321e57466d045d21f1d285519b5d12ae47bd"} +{"id": "train_044095", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5c4327c7fc2a811a6ace27580e4df3f6a653fea1"} +{"id": "train_044096", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a0a1130e17e59b92b3702cef2afd226310eaede9"} +{"id": "train_044097", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ee19a468ce50ab3e5df143eaf0c65171feb2e0fa"} +{"id": "train_044098", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f6e5b9238bfbc8f35840cc3e5fb35589925d1d17"} +{"id": "train_044099", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "645013386f862c2ae542f049cdcf94d7d0655a26"} +{"id": "train_044100", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "4464ab23085af6d3bbc9854240182a4af7a6fe44"} +{"id": "train_044101", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "cc300882ab2ffdfff476bdbfc09540fcb1330177"} +{"id": "train_044102", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0de7d928d243dcb27ff070283ac3a520f8d8e91d"} +{"id": "train_044103", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ac224a2cb01fc1bc14f44dad6a2f6022d1015e8b"} +{"id": "train_044104", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7ca808b7a82b0e0ebd0cf1c42ce52e3999833392"} +{"id": "train_044105", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a665c7853ea1cb16bf5785721d9723e81341c9c3"} +{"id": "train_044106", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ddc0274a1f8360c842f9d339a04fbb2d50444b9c"} +{"id": "train_044107", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0690e629dd3fe259862a63dbb3cb156f6d63072e"} +{"id": "train_044108", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7421cf5b0ea1f60edaffaf168c5ae80715be7c1b"} +{"id": "train_044109", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b3ddd2b89f0d0d1435731f7450ea95dcc8439796"} +{"id": "train_044110", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "787dd58d6332b6ce8db1adda40d50c67f92e7fe4"} +{"id": "train_044111", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a634b7eb96cd809556407b63f10542762675989d"} +{"id": "train_044112", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "72e7ca0b393fcf81d7da4ea70d56c99a75459fa9"} +{"id": "train_044113", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7acbd9292c3663db247ad9e55047bfcd3ecf3fa1"} +{"id": "train_044114", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6b1018b4be3ff30d1af5b70dd896a2e87d021b94"} +{"id": "train_044115", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b42e994c4995c69f772a063e63a4af94a9c7a3aa"} +{"id": "train_044116", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7d6e3e098a7b970e12aa15545900adb35bd02c0d"} +{"id": "train_044117", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7b72e195903a705eace20b82ef4748a712da0fd0"} +{"id": "train_044118", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "321b2e5cdd2cfd47828bd70c1684f342871b049f"} +{"id": "train_044119", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a205eb1412c2766d3fb135eee303eda177cdf7f7"} +{"id": "train_044120", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9f855d3f2165952258b989584036cb7b135114bb"} +{"id": "train_044121", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7c07146d0c39c737f9401b285ff9ab84e93402a4"} +{"id": "train_044122", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "55d876613fede9d4611d53ea20e23d38ab40bbcd"} +{"id": "train_044123", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "4f8db433d445fd38ea32b6dd9ab1bbfc38f07197"} +{"id": "train_044124", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3c365f2ec07542845af14d390290f33eee92a0fa"} +{"id": "train_044125", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "36ebdd643b7674e1bbec6b2c2c6227ebc20e63ae"} +{"id": "train_044126", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c04a274bec4d1eea6a73755e99c496c43f2893c3"} +{"id": "train_044127", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0da11b72f051a1906a1a67597f409db8fa042a61"} +{"id": "train_044128", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "cd4b20f4e90364dc636991431bca3de3550c74a3"} +{"id": "train_044129", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f577fa16b70c1d5ffda5cf3f6a98aac50e4ef8e7"} +{"id": "train_044130", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "6f2be56184e62acf441771e72381fbb58dcdf263"} +{"id": "train_044131", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d52fe97b6c56555336376ada9ee8b5d9e88a0648"} +{"id": "train_044132", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "e2bab2f8163b0b554833952e35fa72793ae22d6b"} +{"id": "train_044133", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "02a29901bf2b3d1e5af04891bab6685c4607c920"} +{"id": "train_044134", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "212aa9bfaf6b97a62695dfd2af5427c114d30854"} +{"id": "train_044135", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "044288feab7378d729bedea61841c74493b07cb3"} +{"id": "train_044136", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "bcd5655009c604ff2b2803228ce62c7f3adeeab9"} +{"id": "train_044137", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "dab4828f5b498fe986e1a17a836345619460b9e6"} +{"id": "train_044138", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "130475008565040bb37219de91ddcde617ac0d03"} +{"id": "train_044139", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "67b0377134816d7f28c3505f53e8263d2b6bcc54"} +{"id": "train_044140", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "50616df38e5e36c02dbb4aa90e285f5691d31a52"} +{"id": "train_044141", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1b94a7850cce5419e1143edf8d609cdeff1a5d1d"} +{"id": "train_044142", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9f74a73b14e403b13de1eb1d8cb4ec4b63be7e15"} +{"id": "train_044143", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8ea0d31afd0a7647dbef7423f01cdbf8d3d35a0a"} +{"id": "train_044144", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "38a33301da710ee8737a234b82ea34bc3600b2ca"} +{"id": "train_044145", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "eb0e8a12def9e8bfa7e13278efffa901e26df61e"} +{"id": "train_044146", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9eb1a449a8f0bf77f47c53fe939572a89ab9e6f5"} +{"id": "train_044147", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "6fb466e271b43ff7ba859bc07a4cc2b5e6e15e2b"} +{"id": "train_044148", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ece07dc8edb94cad198d6df36d8ab30f4b7f56f1"} +{"id": "train_044149", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "15a499cf24c00c94b35f0c7367e0636d1d9be2c0"} +{"id": "train_044150", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d570eb4eee097419edcece127c710da099bf247d"} +{"id": "train_044151", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "53562430d9d5a3fb65803433d143c3d220ae8bfc"} +{"id": "train_044152", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["research", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4fd77079a6337364ba3053a6bac69687b28127b4"} +{"id": "train_044153", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "77aa7b9e17395cf19bf1cd38bb424fce2ef61575"} +{"id": "train_044154", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "39707bc5e84a98352b84657ff08e85560790141d"} +{"id": "train_044155", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "fa8698f2e37c3a6145dc7834630bd8dc35a7ece9"} +{"id": "train_044156", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2d6dd54d9cae93ed34dc29b4c2760e7ef915a0f9"} +{"id": "train_044157", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "cd6eed813887c2ae14926383bd7ffb718a916e95"} +{"id": "train_044158", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "50d0e2a0a4fca380203099e82c04651d203a66ca"} +{"id": "train_044159", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "05718999ebca307ffece29b0fe3970e7c2eadfdf"} +{"id": "train_044160", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "24a1b4473961548e1e62eee21420cf1aa6535c92"} +{"id": "train_044161", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9202fefa9d29f3ab6f88eee8f1a4a7c699d51e2a"} +{"id": "train_044162", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0a4d1c125c9fd25f72cd7d5db32a91cb4b84a895"} +{"id": "train_044163", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "8c4c8e7000d98e38ae32883ab5b24c4fbc89f689"} +{"id": "train_044164", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "76ec744f4262c33e4242b3f6c7600cabeb76480a"} +{"id": "train_044165", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "4013265145ba93a56ad6890db0c5016d13e4c3c3"} +{"id": "train_044166", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0baad7eecd7305a3341fe75cce4b0753af6169be"} +{"id": "train_044167", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "518ad6bbe761f532e1faa800c652ad18e08d3c9f"} +{"id": "train_044168", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1ef0506e100603eca5ccf97caed621512783a8e5"} +{"id": "train_044169", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1d694b870738360b81dc8b19ae2b3fb20e706eae"} +{"id": "train_044170", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "18142e92270d5d9f84c82975ac68daed151c01a4"} +{"id": "train_044171", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "763ab5581622813ea10cceb3d599a7638aeb4daf"} +{"id": "train_044172", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e374adf6abb61cc25c047ff98e4c8a5a8e7cacee"} +{"id": "train_044173", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "15ae73004dafbb87ea1cc49fcd3119de8f45d3a5"} +{"id": "train_044174", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c973c1b4bbe813839b702164c5638a1cb9c065aa"} +{"id": "train_044175", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7ae569fda34761fadca28b35b1378140e819d81f"} +{"id": "train_044176", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a479df8d0f6888a05e07afc639e0cea241505a32"} +{"id": "train_044177", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "996030e653deaf18ba537ea37f85f47aacbf0af0"} +{"id": "train_044178", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "b280ac9241e95a1b3428a527f22138677b720754"} +{"id": "train_044179", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c7e6f6eb9695350371d32021c651c187dc2e58b8"} +{"id": "train_044180", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "01b367e67a890d159f095205bd95bccd9a6115d2"} +{"id": "train_044181", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2a10bc6a89c998d3282bdc56d110ad0b4bcf1781"} +{"id": "train_044182", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5c0cae619ebdd32f8949826d1a1f2a0e455bea9c"} +{"id": "train_044183", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "16295dd47e33c1a083935ac043c3f3434a60e4a1"} +{"id": "train_044184", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b44f81323d2cdc66e23b1e6c449d416457842f04"} +{"id": "train_044185", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "da9b942637130ed1d3953cb6b076279c00f8282e"} +{"id": "train_044186", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "1ac95a99261302015ece116b933e4ea70d53303e"} +{"id": "train_044187", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "6043a3c28091f5956c67a4b79bce596a7eff8a72"} +{"id": "train_044188", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d597258fa4d26bcc2bf47e33c63b33e4198fcfef"} +{"id": "train_044189", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "15953359d4a5781ca3674d9bb7b592368ea22633"} +{"id": "train_044190", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b8dd95467c58f8093929d87c431bd67abf4ec173"} +{"id": "train_044191", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d69a8ce6aaf9762ac2834ddd2947604c4aa88da4"} +{"id": "train_044192", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7fdc534b4da63b78a2a44195aefa3ee0cf76a1d9"} +{"id": "train_044193", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e91d3aaf89dfc3a576f105bc7010e3b1a9d2e506"} +{"id": "train_044194", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "39349d4a8c5fb334fe0ca9f56dfe7b9ab8e682a5"} +{"id": "train_044195", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0d964c9c3a9e09967d94615e634cb3fc096821ef"} +{"id": "train_044196", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "57c05931614dee6b33e6a2c1709c253753aabc62"} +{"id": "train_044197", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ce75a1c46cae76df0549bd4e237bb6d199b8d7f5"} +{"id": "train_044198", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c7041f01c74f1d6fd6f260d16be32b5b4f1a0eb8"} +{"id": "train_044199", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6b42a98f8c62d49aa6343f605305598039d3152f"} +{"id": "train_044200", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "35d792721e8c432d2f935dfa6f42b432c46a545a"} +{"id": "train_044201", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2325faf740603a82df066d181fb1a314308c662e"} +{"id": "train_044202", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "161286222720cd874f565bb8dc5894bc674e8b0a"} +{"id": "train_044203", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "efdb96a0e3d22bf887f6584081adc4b5b9e1fc6d"} +{"id": "train_044204", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b660db621d1cc63867939575e15c1d6c41732dc8"} +{"id": "train_044205", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9d44b0595ef642771e4f504f4cfae5e05c3079c6"} +{"id": "train_044206", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "dae3c9b7df9793876c0f489ba65481c1893784f6"} +{"id": "train_044207", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "20e8766b94a34efac577dbfda7f8fd0aad80ebf9"} +{"id": "train_044208", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2d609044a36b583893dcdb6ad607b2710d65824e"} +{"id": "train_044209", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e3e281cbdde874dd0b0f13b5559cc0d8fe4d2713"} +{"id": "train_044210", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ea258c470a09d7ce47c30cf5fd7369f3e35bfe84"} +{"id": "train_044211", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "dde1728acbeab6f6d23e57363d2e894bbbe32cbc"} +{"id": "train_044212", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "714d6cf8b17f67ec52563929d94a01bdc3f1adc7"} +{"id": "train_044213", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7f6acc6c0acb1ba4ee8798d5d39bd9c9967117a4"} +{"id": "train_044214", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "203a7e4b8ef4e334358576d7a6e3b44ca2c9a8b3"} +{"id": "train_044215", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "96fb9db1114af49e0bcc2e4e1cffb7420aef9123"} +{"id": "train_044216", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c1ad027fda84229d6d51292255aecdb81f97bc2a"} +{"id": "train_044217", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "2ed21fcf82398c918afa3ed0e6dbbf0244c8666a"} +{"id": "train_044218", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "eefd915c00b15604ab030238ff11a87bd12c6731"} +{"id": "train_044219", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1eff519f967f72576d15f893486108503aa6fbd2"} +{"id": "train_044220", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a40733a05ba08c8c0208a36c94b59b65a0a0fb1f"} +{"id": "train_044221", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "553f866c49cdefc0acb55fe4184c3b3da4808365"} +{"id": "train_044222", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "cc7e5d5f30b32bafec946cf0c6057b007a844519"} +{"id": "train_044223", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "648218a315ad11a991c55b8c7bf59b6febb41132"} +{"id": "train_044224", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8bcf3e2ce945b07fd7de016e369e41b2bd020706"} +{"id": "train_044225", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b0b1b2490061d987e1db17b3b3537c8bcc40e278"} +{"id": "train_044226", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2ca6310a8e70e1f721611a5354362e9add1e1748"} +{"id": "train_044227", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b543b63c7ef030357cc40f653a60cc75a48c1b2e"} +{"id": "train_044228", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0a0d1b32053a63163d38c55ef45483be35c4173c"} +{"id": "train_044229", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7c65efebc8f02b49847e44aec0fe3ad854c04744"} +{"id": "train_044230", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3bb3a2316527668e0984c0f9e4aa6dca750040aa"} +{"id": "train_044231", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "23e8c4ddcfee7779193d4edab6a7e267ab021373"} +{"id": "train_044232", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "07947214b6acf2745cb174ec40a79da2b2bd12d5"} +{"id": "train_044233", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "fcaf58a302311f7daf2270fdb265b893100c4e47"} +{"id": "train_044234", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a05fcedcc8f4f4cba36dd37d4a1919aa92e34ff4"} +{"id": "train_044235", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "435bfe5ba13aa9877a01dc589ab46ace720b62ab"} +{"id": "train_044236", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["data", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "eb5e3346b6b26c6d2b69bc9510e89033c6973d59"} +{"id": "train_044237", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c669a81f11cd84ebdc5a680cacd9c6a9d6739a48"} +{"id": "train_044238", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f899c28e1391de17c0a47b8690b128bb4e0352c1"} +{"id": "train_044239", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f91c1a1e1ce0f692d7a971d8f146aa097fd76cf8"} +{"id": "train_044240", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1af55fd81195c1b1cfbe7a878b142008f8d23f56"} +{"id": "train_044241", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2f54caaeccbce50cc3b4693e3f28912135195f1f"} +{"id": "train_044242", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b31c58b4fe07315fe98954eac56edf161d02e88e"} +{"id": "train_044243", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "cc7e5d5f30b32bafec946cf0c6057b007a844519"} +{"id": "train_044244", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "22de7008d7a4a9decec496a39540f35631194cce"} +{"id": "train_044245", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "11800fe3ebeb4663adf80d5fc405e8295f3b65ac"} +{"id": "train_044246", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b20b29ee798b67f99e73dad44e4c4ce406ae9b47"} +{"id": "train_044247", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8d9d2d24abdf50f827584a0f2134d5ee6c63ce9f"} +{"id": "train_044248", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5c3983c985d8507c380ab78f97ed982e712f7fc0"} +{"id": "train_044249", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2bedfbf83787c6c05e8bee77c8e5d6fad768f1b7"} +{"id": "train_044250", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ef9cc0cffd531b25ddfb41b2a18b9d9b5acc5c31"} +{"id": "train_044251", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f334495ccc8a732ba320c4bb8cfa9bd0dde56663"} +{"id": "train_044252", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "95436c3a24f8b2e82249525082bb88297cd67ee4"} +{"id": "train_044253", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "144fce9834737c201451823caa0aa28cccfdc818"} +{"id": "train_044254", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "7588740cad867d7fd3f1faf2a510486fd782aa05"} +{"id": "train_044255", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f2c20d2c67fcf4c14db4ef31acf1267c203caea3"} +{"id": "train_044256", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "5f55dc3f44aae9963d402720be14566348b36942"} +{"id": "train_044257", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a6df04280387e266ca50cc340b2b2bb85d2fffad"} +{"id": "train_044258", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "98565a359e621c1b5649ffa448874459bfb33072"} +{"id": "train_044259", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6d073ee1e6e395f3474cb6d94062f704aabcdb16"} +{"id": "train_044260", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "23300a9890c815a3d672c6a4069bf6194d5440f7"} +{"id": "train_044261", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8c60bb6a907b93c8b7be7aa7373db9f9283fce27"} +{"id": "train_044262", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "71acde6ddd5a32b46e04d654c0512788b3235dde"} +{"id": "train_044263", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "da287370cd4be8a69b94240c39c90493f4d986d7"} +{"id": "train_044264", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dc20d822792c9fe4eb7a0ea4ef6e68362c8aea19"} +{"id": "train_044265", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2cce4b83360a7511c48594fd82fdf9af00943dd3"} +{"id": "train_044266", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "4f8db433d445fd38ea32b6dd9ab1bbfc38f07197"} +{"id": "train_044267", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b58017e5eb0ff2384dc3b031b397a8f8063b1570"} +{"id": "train_044268", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "35ad17b7911bb526d39267c17a1bfb8170450dc7"} +{"id": "train_044269", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "eeb7e76c0035f3d3ca1c27bf3809fbc46a7c6da0"} +{"id": "train_044270", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "20e8766b94a34efac577dbfda7f8fd0aad80ebf9"} +{"id": "train_044271", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5f076c2a5c72c08c3741f5666d8a14ae2068b309"} +{"id": "train_044272", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e6cdec9a5418dceaf8645ebf67dadf8e99f34547"} +{"id": "train_044273", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "cf9d8a803b5055967a1c16ac4ffd69367a3cd6b7"} +{"id": "train_044274", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7c34ffc1307987364d467d3ceb7af9bc9000e6bc"} +{"id": "train_044275", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "21ab7502d52ec404152d0c5bc5af38ebfc4fd41c"} +{"id": "train_044276", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2f7ec88138d2b709b9d0c9ace328cb8a05cd04ef"} +{"id": "train_044277", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d2a89726f7ac317902cdb8b811a1f5741e7bf12c"} +{"id": "train_044278", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "62aba1337de31da34597feacf49d48a9b421a8d5"} +{"id": "train_044279", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a2fe3cfea76edeaded90edb9768366d9e3783a5c"} +{"id": "train_044280", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "966d03d39e8482a7e6932a476f022378878ea637"} +{"id": "train_044281", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9aa31ba152b4fce020b1273b4eb16671d4b0a6ff"} +{"id": "train_044282", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a44db2c6d3dbff7ac7d16a6e6751d2d5ba4bd322"} +{"id": "train_044283", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "133b8c60eff95201c977079c5dc6e4463324a8f3"} +{"id": "train_044284", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "50fdb92db0b7fc2ce79a40bae8d665fb7321ccef"} +{"id": "train_044285", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "9e69cedf0d37ab64439ec5205d62f8eb1442b056"} +{"id": "train_044286", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "9b7b726e1b6bd89d84a5a30dfdfe120cbf759824"} +{"id": "train_044287", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "fdca3cbf8cef3cf7d66fb5a96ce4e3115b97cd13"} +{"id": "train_044288", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "11522dda8aae6e31bdb3ccfbde0797a48e1090a0"} +{"id": "train_044289", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "634ee81623093658004fd1f92a2997496254bae3"} +{"id": "train_044290", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "fd463cd19762653d1cbf564fa462118e02398407"} +{"id": "train_044291", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d1c28e59a0240d75bdc2b5dfc4b717601c991e06"} +{"id": "train_044292", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "633d8f44c0f76a1715216206b8d9088219e265a3"} +{"id": "train_044293", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8cade80496aa66e910361e277c78d3f3b0d90ecd"} +{"id": "train_044294", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1a9de3b1dc7f04d141576af071b71c903670da0b"} +{"id": "train_044295", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "592c9ac371ff11ffb7f0db0a3d4b7b3c1619c4e1"} +{"id": "train_044296", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5594c63b20d042d4d6e976e3b633c9c16d817601"} +{"id": "train_044297", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "0e0b34c725f0d6276faa6b1dd14f73cf0ae46348"} +{"id": "train_044298", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "182870542e3ffe2cadba5c75e0c7895de77389b8"} +{"id": "train_044299", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "91c4d658a1f49f72387cc183080b57411bc708e9"} +{"id": "train_044300", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "218be2544a84c6489cdd34e65a767c9fe08519a5"} +{"id": "train_044301", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d0e4f7885824c5f666b8e2d52753ca12111d8368"} +{"id": "train_044302", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "80fda4f72739960c1a792becf58c1a84655e4919"} +{"id": "train_044303", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f9a079539a0cba336b7f02db44c3683ba4933965"} +{"id": "train_044304", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "298e2ca2bcbedc4c6c7a855aadc314126e548efd"} +{"id": "train_044305", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "4475b56f6ddfda3385da58e800dc4336c13d5759"} +{"id": "train_044306", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7e43b7029b231015d6eeec1e1a199ceca6826472"} +{"id": "train_044307", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "8d5523318ca3b8f4e6a3d04bd627045720185fb8"} +{"id": "train_044308", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "c87c3c591b623eeef946540f973892d5bc16f3da"} +{"id": "train_044309", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b08c46236ad736d348662ded719d865342c72a40"} +{"id": "train_044310", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "da94e7053057ab3b8a715bdf98cb05a229977090"} +{"id": "train_044311", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "462d9b154e12a6508aa1008ab3e2a4fd375fbcfb"} +{"id": "train_044312", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d2ab8ff50a6f0bebf51a53e681933cbc9a3235c2"} +{"id": "train_044313", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "38f1a0abe1ac4fd08c77a2127de90867e225a4c0"} +{"id": "train_044314", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a97ae3c3479ee46945e903fa9e617fc96f3a46a9"} +{"id": "train_044315", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "29b41bc8ced02bcadb7d10de0d1c417d928c47e2"} +{"id": "train_044316", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e699fdf6c88d4b91fb3f67152c72bdeef0e08819"} +{"id": "train_044317", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "29429b1a1662532460d89455ca9a34b930be2378"} +{"id": "train_044318", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "065b85a51c37d2d2cd083ca34ca8f3c287324d1f"} +{"id": "train_044319", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1569936c8970112ce1e299ed74e09ca027fa38f3"} +{"id": "train_044320", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d87d89077e5850e1690751bfe8cd11b111a2daea"} +{"id": "train_044321", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f16cf988ac4d90930177611eb0c17d867125cd42"} +{"id": "train_044322", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "abcbbf2c0429ffa3094f394e2589e58dd1c0ab7c"} +{"id": "train_044323", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "187ef2462c6b6cda1d739c0f14fd2c9156f78dd1"} +{"id": "train_044324", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "bd1fd3cdbc932d0863bb4c120fa47e315c371237"} +{"id": "train_044325", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "6a934f6e7a23f8c1695128d560564bebef6cb35f"} +{"id": "train_044326", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "fe004da59b82425f36f5f7be4a379d72af7b69c5"} +{"id": "train_044327", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8c6edcb78ff09f75800dd1cda4fe9dd1f2f12510"} +{"id": "train_044328", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8638dfce1e9dde4c4701b122a09c32714be5dc5b"} +{"id": "train_044329", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "340a751b4d7acf2ba1702c8d02aa6b824f3d1a6f"} +{"id": "train_044330", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "384cb83d6350a86fde6a5eff93d85dff8387f76d"} +{"id": "train_044331", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "767bf58ebd0b1c3ce108c0892d5ed79229334c17"} +{"id": "train_044332", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["agent", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "02561af755791af61118326abc59ef9c13e56643"} +{"id": "train_044333", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "be5f754434c5882c7e44040caef7f36d4c221057"} +{"id": "train_044334", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "28c70a8df8a3c0376fb794ade82d6ff0541db91f"} +{"id": "train_044335", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "962ed4a60f3d11ed6778babcbcb5582f3df75bd8"} +{"id": "train_044336", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5de52ddf3194776a11dc0c21238af88edda93100"} +{"id": "train_044337", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5d4315dc9f90711b9b608b4d5c09db81f38f491d"} +{"id": "train_044338", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b1fadd6aaea481f521da88d8b72a97ce5d0cac67"} +{"id": "train_044339", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ab44433d87f3f2bbdedf48b7657f0a747c5a3938"} +{"id": "train_044340", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "027eb0d7c4e9f4e171be0ca359c8b211a2338ba7"} +{"id": "train_044341", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1a082629ad138bd2f4e83ed4bbb7d5878d46e938"} +{"id": "train_044342", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "50db4a4080827dbbaa4ee38b8432d5ce3141f6e7"} +{"id": "train_044343", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2f888f9deee276cd999854ec29a9c373a52d106e"} +{"id": "train_044344", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "827ac587ec6d47e1d4c56de18efe270c11b34290"} +{"id": "train_044345", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1804e522f1e7b425307b510013652647318433c1"} +{"id": "train_044346", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "5c3983c985d8507c380ab78f97ed982e712f7fc0"} +{"id": "train_044347", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "7915cc497abff575246d1e3256dd1fbe93df8b8b"} +{"id": "train_044348", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "38d0c53c3a77e66328675f021a4391ab8020bca1"} +{"id": "train_044349", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "0ce0595ae669ddea7fa0851bef555f69064387d3"} +{"id": "train_044350", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "42bd266260b033157c1aa179b18e5774f7b8639b"} +{"id": "train_044351", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e4fc9fc2edcf22f7c8666ba0fe641935d9eadf98"} +{"id": "train_044352", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ec6e5f8d760db1d14ca9c8ac441feba98bc963e0"} +{"id": "train_044353", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "80cc2eabdc9c8f8a05eab9fe5c6f73adf0b3fc0c"} +{"id": "train_044354", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "385c27bf799437741f82191f11767bd6bf9d073a"} +{"id": "train_044355", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "602b305216cb0024700a32c0ab1bcc85479de2aa"} +{"id": "train_044356", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fdece1fcd2d38ce45372d27ae64b3b7101cfe39e"} +{"id": "train_044357", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "240462aaf8fe71b56565a10386e39eea42511e19"} +{"id": "train_044358", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0bad4323a0af4e0ddef61538cd48e7031ca8f76c"} +{"id": "train_044359", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "389f69db95cd553ce9fe5658e8be1f009aabf5f5"} +{"id": "train_044360", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "175a9e4f319e89d6c8e3643d89d52692949bf0c1"} +{"id": "train_044361", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ee7d10fbb60764854446db025361f9a115f220db"} +{"id": "train_044362", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["research", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9724390f08935501ca487aea888ea8f0da78de4c"} +{"id": "train_044363", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "643b6fc717244e138a6492f76a7c1fb70268ad53"} +{"id": "train_044364", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2bedfbf83787c6c05e8bee77c8e5d6fad768f1b7"} +{"id": "train_044365", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f8622622280a5c62bf9a0cb63df1a0e0b4ebcae3"} +{"id": "train_044366", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3351845424ef58085749162768503f51829f7582"} +{"id": "train_044367", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "16884e3c1d8c209b4a5ac561deec940c48769f78"} +{"id": "train_044368", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1cde996d8a4ac5f366b6978db318cecc440e0a2b"} +{"id": "train_044369", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "9a5a6aec6f94d868b08aafe62df2f0217eeac9db"} +{"id": "train_044370", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2b6d4d820d32022f1f9d26751c4e8e1c6ba32294"} +{"id": "train_044371", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "412074daf688119b150ea60e9da3c4bda8eebd82"} +{"id": "train_044372", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "09ef97a36071b21cfa2517161991990e3c0a2092"} +{"id": "train_044373", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ffe2137e3fa0a78adc1df8423379946d045328a4"} +{"id": "train_044374", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "ee1b0fe37730cc0d34d9475b8472eb5305464b72"} +{"id": "train_044375", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e3b196f0e76d9396bc812f10e999717f1707a948"} +{"id": "train_044376", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "187ef2462c6b6cda1d739c0f14fd2c9156f78dd1"} +{"id": "train_044377", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c85eec53c81a657bee363b362bb312988709e949"} +{"id": "train_044378", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "fd3e18d344c96e6df6e6f9cd145f4b858fdf804a"} +{"id": "train_044379", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "0ef015aae168b3f9f5f6532bc408d8d4b32f869a"} +{"id": "train_044380", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "79679a2565d38789e0e7452078cb4e8ec34458a5"} +{"id": "train_044381", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1c803d7d28c9da7d874c41a1048be3a2bcc280db"} +{"id": "train_044382", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ae6319b29c3d21b0d9e25048fded357f17e24b48"} +{"id": "train_044383", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "cdca2f1b9c93b8c81022fdcc4dca4fac44f09388"} +{"id": "train_044384", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "96c8a826f15a260305845f2cabcd0d90b8bff813"} +{"id": "train_044385", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c6cb3bf4c5001c6a2672d63c5a839dbdab06b858"} +{"id": "train_044386", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "bddaab2c7d8e6be63bdbd16d2eee468ad1e72954"} +{"id": "train_044387", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0ae48054c05b7f92049dfba1f9fe8efa199ac02e"} +{"id": "train_044388", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "c98d0ebcef026083f66c93af9c017c27be40fc9e"} +{"id": "train_044389", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["data", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "99aee1117507ad1bcd7c8d315581e899313c63f4"} +{"id": "train_044390", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0c7fc467367c957b711343eb47954707b3c258c6"} +{"id": "train_044391", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ba8a8200a7ceb2378245b6c19d0bf2635454b6aa"} +{"id": "train_044392", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8b47ea968a573c9649d3b02145b8bfe5c5de3446"} +{"id": "train_044393", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "bdce22dc73ecc87e8c8b2c810c183a08112be242"} +{"id": "train_044394", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "ff6bf48bb8c6a63444d0005e1823ccc599aff13e"} +{"id": "train_044395", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "fc8172e5aba6ff1afa58559d91bddcc4e1e60802"} +{"id": "train_044396", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "093e8b4a0f49948553a5b303d5d4b759f5ede253"} +{"id": "train_044397", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4369f8c1ed21efb15d8d645c98bb6f5ee90341d2"} +{"id": "train_044398", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4e341022d381dea5ec1e219e390459b39e02ebb9"} +{"id": "train_044399", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1e37e49d8f7c2cf54759bd7c0a0172247e8872d4"} +{"id": "train_044400", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ff6ead12fdc96534d2c6c91c543d73b6e6fc5659"} +{"id": "train_044401", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "464e07d102e012419e42235503cdb7c6299f748c"} +{"id": "train_044402", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "9859cf2f0a98f6b1a3db1f6c508b00087d421307"} +{"id": "train_044403", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "913262744759ae110d31b2c8b7f759ab5a66ebec"} +{"id": "train_044404", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ae1875a74e9c2cd933835e6a793801d3dc17412a"} +{"id": "train_044405", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f6f0e445d6c09b01c3837928161cea31f03f5830"} +{"id": "train_044406", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3ee04c0a914b962ef66485db9f1de5f7178ad63b"} +{"id": "train_044407", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "3a9e996bdae2bc0af778ad207bec77b726261c6c"} +{"id": "train_044408", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "c59c33dba3ac0700ba0142bf8f3c483a259b48e0"} +{"id": "train_044409", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0735a754224c15b9939cd70d5175159188f1f919"} +{"id": "train_044410", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "72c3e5e37df11a2850d034fd6663e6059d00d26a"} +{"id": "train_044411", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "03a846664de842c11bc577767d061582b8682f93"} +{"id": "train_044412", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b09dbfadea18516cf79f76d345b9a2b0937d003b"} +{"id": "train_044413", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bdc384b230090c50841f7b05aae667bb85e138f1"} +{"id": "train_044414", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2904d0f32a3d85cdeda486400a6c6925938cc93d"} +{"id": "train_044415", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "30b5d97926c8b786510b91489fbec8706fe47fd5"} +{"id": "train_044416", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "186e01f5f5b72b55bf59735daa70f3b1a707bce2"} +{"id": "train_044417", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4daf6ae0e0be5fc793e37654f8236b859e28219c"} +{"id": "train_044418", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "02cf08cfa45520ab9a44928eecc6501a30cedccf"} +{"id": "train_044419", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ee86bcceb079990cc9efcea024c73c2758a70139"} +{"id": "train_044420", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "de7d70cb36773a5c6b51ce913943256c78671ff2"} +{"id": "train_044421", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b7b1b03cc1b9b13c9603deaa70f7e02a160a8628"} +{"id": "train_044422", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "23b2ce3f615c23d2d0394e30903f2532792eb354"} +{"id": "train_044423", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "378126deef11d8b63402d63a695ef78b27f1573f"} +{"id": "train_044424", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "189e1489df6b93c0b84a5366e439ea734d4e67eb"} +{"id": "train_044425", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1f9dc8759ba45c2aaee65c9bc6d7ebb6d1e535f2"} +{"id": "train_044426", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c932fd7a65f0f0fb2aa81d2d434dd5469a3fa8ff"} +{"id": "train_044427", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a12e7e9a34c0538f1ba47bb31419b1ef2ef6a2d1"} +{"id": "train_044428", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "73e6657c630cf2ef65e46d6950ea096ae211bb79"} +{"id": "train_044429", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2c034c64c0c4c206b0481b8a2ac9f0bc593c5ba2"} +{"id": "train_044430", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2bc3cf10648a9df1fa3b36f65b502fe59161431b"} +{"id": "train_044431", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c0035cc91d5a4807a81610cb21e54e6ba230d995"} +{"id": "train_044432", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "194d2e90a513ce082cde373d29f8057b0ef0cf04"} +{"id": "train_044433", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "eac6f4abb6b6154a4589ea13e332bcac0cd0b65a"} +{"id": "train_044434", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "4af37fd92dbda49d6ba864bc5c4fe76de03d1e86"} +{"id": "train_044435", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "45af494ff1117d6ae7bb016728047d959f35aa3d"} +{"id": "train_044436", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6a0239bc8b675fdc37d8022d1a8aa2dd25426267"} +{"id": "train_044437", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e01a6a2b91f5d8ea2a3f253b4b37bed7e187ec4c"} +{"id": "train_044438", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "06668b90e9bda575851b588463a6dd97daebf1dd"} +{"id": "train_044439", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "dcfa9c92bfa9be6e92b4c1c9d5c0d78095457502"} +{"id": "train_044440", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f7ab5028d9a60f30014e79586299bb02387dc7fe"} +{"id": "train_044441", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1b8a73c7523f5696eeae40f76e57341c3dd1fcc2"} +{"id": "train_044442", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["performance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a00d2db7785982faa680e4101acbdfdddf107ef8"} +{"id": "train_044443", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "7de9949b6bbecda656640b3751195462aa823c76"} +{"id": "train_044444", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0302afcbca45d4961b34a940726c0e8f52ae1472"} +{"id": "train_044445", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "55ec7f903437a3f95c5923c1b29547bf39e17b36"} +{"id": "train_044446", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c15733263e0d701a8e6617a4e9f857f83a17af27"} +{"id": "train_044447", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "74ac7b7020f97679a416bc3c9ce50a7c54010471"} +{"id": "train_044448", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "5c5b9b5fc9002bf41bdc6134e46b03168ab011e7"} +{"id": "train_044449", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "dc29ef4e6ab311b1812cd064024aea43447b96d2"} +{"id": "train_044450", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a5d6a077ae8b7df04ccd68a476cccba728adcc38"} +{"id": "train_044451", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3f5fda589bf383a7942d0be4101bd72db4f84326"} +{"id": "train_044452", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "71894cd5a36997af87b8ea42bbc288814c2b68c2"} +{"id": "train_044453", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "291a62a253c0098699356a09cce1b517ca105ed5"} +{"id": "train_044454", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1c1156078237ddcb12f01d78d680f5574b6f969b"} +{"id": "train_044455", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2bcb4caef884a0c35ff842bcc92d71fba52093b6"} +{"id": "train_044456", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "25a030436759f9749221c843665e95566b4dd6a1"} +{"id": "train_044457", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "7dcc407b97997a41df05e0c4747720e7fae01104"} +{"id": "train_044458", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "6a028cc6a11c6d265cc9e7880cbec96c69173683"} +{"id": "train_044459", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5bfca95102c26c9aaeb8e38d5d12246031412cb0"} +{"id": "train_044460", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["coding", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ac4781e10faa3b1ce6bb4ea2fbdc10665191061d"} +{"id": "train_044461", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a54e34fb27d78c48d941f52c1f49d98398c2c270"} +{"id": "train_044462", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "65b84141c6782846d63f647eae54d1350062ab78"} +{"id": "train_044463", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "161286222720cd874f565bb8dc5894bc674e8b0a"} +{"id": "train_044464", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ed6746ddaa8799893aa0ecc9322f0e062330333a"} +{"id": "train_044465", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f784f2960db226a72b7c5e63c0b9fc84d544d53f"} +{"id": "train_044466", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e6d30346f2e11eacdb28bc83557985b2fd2a54b0"} +{"id": "train_044467", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "35db3d223e108e90fd6f8380e6a2a19319e96c4a"} +{"id": "train_044468", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e2eaf6200a1cffd9a4ce9b841321aba2232c348a"} +{"id": "train_044469", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b3dfcc143d09174b0d0b06ef6d2ebd72a5b6d533"} +{"id": "train_044470", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "00a039d8f2879e5014d976966c4d424eb1bf02ea"} +{"id": "train_044471", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b7ca3cf22067d977c352acb6f837c8259811bdbb"} +{"id": "train_044472", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "25f1329e12cccf55c6120673f87321a5592124a0"} +{"id": "train_044473", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "5bc762f153589aff60f5b8105c72ad7d98256507"} +{"id": "train_044474", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "347b578b984179b0c890c36d38373dc896dd80f7"} +{"id": "train_044475", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "fec61aebc519c1fb9b3706b880623cd077075dbd"} +{"id": "train_044476", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "a6a9eb2f99899e2a218b887cd8e57d9322ddb8de"} +{"id": "train_044477", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "07ee116bfbc28a9166452d63f9ddcd2183243b83"} +{"id": "train_044478", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a0bba63280841063c47f433dff0f42271ee9beb6"} +{"id": "train_044479", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "1a883e1cbd77479c6ddd74463d9fc5c23a95aeec"} +{"id": "train_044480", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "9edffdab2dfe17bdea443c6b673ad105cb23ce6f"} +{"id": "train_044481", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c0e60f09b230333ff5ad47354e65fd7ad9229e71"} +{"id": "train_044482", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "3dcf5735a6938fd0c4f5be429208516d22997a0a"} +{"id": "train_044483", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["coding", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "65b25c00479f4f6869b22a3bbf95ed6239fbd819"} +{"id": "train_044484", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "7ad064ccce055ed79359b3d69fc21a2e534b6ab8"} +{"id": "train_044485", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5c3875792a6d83b956a0b893b21e94191b17c578"} +{"id": "train_044486", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "62758fc927b1f40968c0b7a6ad41dca98d415c1a"} +{"id": "train_044487", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a935896a481785a3e926c56d1f8d0e041a62d126"} +{"id": "train_044488", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5aa272da6c06ce0bf9562041bdcdae44661424fa"} +{"id": "train_044489", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "8cc0e8ecf321b251f439bf95f8ea0fc88b4045c8"} +{"id": "train_044490", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "61861124427c7c7bc46144e74218046c5f68bfa8"} +{"id": "train_044491", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f1705c3afe9fce8538e268a86948df5956892cd2"} +{"id": "train_044492", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "3477ea813843733491046f8a82f4366a2e6740e0"} +{"id": "train_044493", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "10d5336b3dd1d004b5fb63177ec1e6567757a0c1"} +{"id": "train_044494", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "844706bfbb855513d5dfd6e4c94fe76876ee13b4"} +{"id": "train_044495", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5ab4a7824b8d50a383208f412f8302cf33c371c0"} +{"id": "train_044496", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ac89b826a0e9c1cc61eec4ce2d6699ddfc114a8c"} +{"id": "train_044497", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "501a795eb274bcd31122ad585fb3c39bc4ec20a5"} +{"id": "train_044498", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9a2536f097203421465dd0acbaf4e971c58e6979"} +{"id": "train_044499", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "adfb667c13d0f63571cb83fb0d562a68d53fe561"} +{"id": "train_044500", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "50db4a4080827dbbaa4ee38b8432d5ce3141f6e7"} +{"id": "train_044501", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d3ee90fa600f4246c390ae04b49acdffa02e98c3"} +{"id": "train_044502", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a218b40b170441aff1f4504fa77948030c2526e8"} +{"id": "train_044503", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "080f48301329aa21f526b499e35eb250359c78f4"} +{"id": "train_044504", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "42bd266260b033157c1aa179b18e5774f7b8639b"} +{"id": "train_044505", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "5094eb5ea2372f382d06f787f2ecc8bb7de8cc53"} +{"id": "train_044506", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0e8e7659cc9ce219e60db5bd0c7f9bae24f78c0d"} +{"id": "train_044507", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "fac82d338b8dd5b6c6c99dfb1ac265c2d7a80974"} +{"id": "train_044508", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "635623c0a53be7c9eb60887761aac1691db21f4d"} +{"id": "train_044509", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e1a2e3b20fba4d6d2490067283360efbacf1736f"} +{"id": "train_044510", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "89f5310c451e4d92e059c13e06f5c9cecf30f2cd"} +{"id": "train_044511", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "51b2220a0e38e388e5ac7c37d9bbb9ebc959709a"} +{"id": "train_044512", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e8e262feeb1884d2a55a3560653a8b45cf0e10e6"} +{"id": "train_044513", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8ebdc9f2eec024de909debe8f2ca458f094c891a"} +{"id": "train_044514", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b7c3ba471d1e25ae1620910fa7e9026d27c08589"} +{"id": "train_044515", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "5d5c4f56528f591947c93289e967ae7288e54e05"} +{"id": "train_044516", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d035dee82db109fc575594ed1e7424c6636dbd5b"} +{"id": "train_044517", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "697eaf47cea15e6783a980363e82208c91db4969"} +{"id": "train_044518", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "950fb29aaab14335e258d50637e4fc96fabeb85f"} +{"id": "train_044519", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "de806bda1c216b4799b693703f9753c429d59210"} +{"id": "train_044520", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8d83ad86ef8ce764ea9af59f3a36e94792e21e91"} +{"id": "train_044521", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "beb5acf4b3237394aa0e7491f423a7cb7fc89da9"} +{"id": "train_044522", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "af44a8f3e16d1706ec58faedfd0f558f9d576e33"} +{"id": "train_044523", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d43387e6ef70a660f45a0cc449cff579c035f969"} +{"id": "train_044524", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2ab7307fb8385154d4d1089e8d432a0d786a4ea4"} +{"id": "train_044525", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "b527bd368aa106a8320746f63f06f4427a4e1ec3"} +{"id": "train_044526", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5516ed0ea400204216756f398212ca835c56f0db"} +{"id": "train_044527", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "f139cad61c6ca7d51a4847ffd2db7bb48f8666f9"} +{"id": "train_044528", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Python", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "0b2527f3797ddf345f40f7ef838c19f143c062c0"} +{"id": "train_044529", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8db65367a5853eea4829effdbd8bc0ce5a2f6bf4"} +{"id": "train_044530", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fe3dd7119fc96afab48e357475d48cd3c7931d07"} +{"id": "train_044531", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b770cb02c441fc6bc45ddc8cb89950d7dde745ea"} +{"id": "train_044532", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "2b78ae7e41d080b2f0357f6da73b1dc9f80fc7dc"} +{"id": "train_044533", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "7f65171d2af83cd3bcfa703a9001c70046922908"} +{"id": "train_044534", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b7405423ee3108db789368e7b35accb658142872"} +{"id": "train_044535", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e9eb41e5daebcea423edd0a7b5c41253704ec4c4"} +{"id": "train_044536", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "ed0ee556cacb7a8f4c2f82d579ccf6d8603f46c1"} +{"id": "train_044537", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "cbd612e6c6caf25940fb3baa90b9f2af24955d35"} +{"id": "train_044538", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "097a054c3e66ba3b74230a252f06549c6df9eb8f"} +{"id": "train_044539", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6b8c6efe6936be64111a63deae6a0b664c67cd75"} +{"id": "train_044540", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9543dc80390a6d7f66ba82c71a8feab3fc3a1763"} +{"id": "train_044541", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5f55dc3f44aae9963d402720be14566348b36942"} +{"id": "train_044542", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ccbb5c2949b17839947b7b48cf66bbb2f527f951"} +{"id": "train_044543", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f784f2960db226a72b7c5e63c0b9fc84d544d53f"} +{"id": "train_044544", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "759aa86a974fa2a7760536581c1ac053fe603dce"} +{"id": "train_044545", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a5d7e47ce92a611f33e0e84f6a7478a62b672813"} +{"id": "train_044546", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "f342785780e1335ceeba1fc7728a3d2abf3d4405"} +{"id": "train_044547", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["governance", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "aa50d448cf60d8960ae89553cbd790e057138114"} +{"id": "train_044548", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ce8f31e33dd3771f2bd9a9830e818894bbe21845"} +{"id": "train_044549", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "319bce6183eb198233d06b8bb3e3d324dd4cf3c8"} +{"id": "train_044550", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4daf6ae0e0be5fc793e37654f8236b859e28219c"} +{"id": "train_044551", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9c6446545f21f0ee0c952470d6f4c8ba9c7831b9"} +{"id": "train_044552", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e5d174644d0e679bcf10c93fcbacd396157ef35e"} +{"id": "train_044553", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0f0cb0d4ee14968dc53474d9b73ff359a955ca73"} +{"id": "train_044554", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a2ce7c08882776fd71aeeab96449e271b4ea9139"} +{"id": "train_044555", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "592c9ac371ff11ffb7f0db0a3d4b7b3c1619c4e1"} +{"id": "train_044556", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "721d9a8481de72c44e9ed3affc65e69622deee06"} +{"id": "train_044557", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c3ddd0276fbddfb7f277871afbfe6bd65eecd2c4"} +{"id": "train_044558", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "9ea75eb9497f6ae5eacb3077ceb40e7514c73b0e"} +{"id": "train_044559", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "825b735ffe52fb51700b34aea03a2486e314132b"} +{"id": "train_044560", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "b1be0c5f917f6042cd8551aa310238634f3f665d"} +{"id": "train_044561", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "f2bb02baf63497ccf1a715caf6e714d78f05b6a8"} +{"id": "train_044562", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "ded33c4506768d04a757e79d95af1431ad217a72"} +{"id": "train_044563", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d570eb4eee097419edcece127c710da099bf247d"} +{"id": "train_044564", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "b6648a3c2387545e8d5e479ac86df60acdf0fe36"} +{"id": "train_044565", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["agent", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "a0a4588d1b30617f882e11011362493bb5212c5a"} +{"id": "train_044566", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9e9e07f06cd54b1a9322f7fdcc192a87088710ab"} +{"id": "train_044567", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3e14721dcfd915628a81f972b399eb8da79524ab"} +{"id": "train_044568", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d7c3a2582823794d3e21938ac2861268c8a9855d"} +{"id": "train_044569", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0007bdcd575c43806e52d8cedb568a7ae32d6ef4"} +{"id": "train_044570", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "feec037d2db42e5d224958f4ac9cc036afff8ed3"} +{"id": "train_044571", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "53505dcdc76f832eabf6f54051c1d500532ea24f"} +{"id": "train_044572", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2081053be27b5b2c6bcff63f0a55994d0e5243d9"} +{"id": "train_044573", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["coding", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7ebd4a59790ae9b4b6e69feea8842720c81242da"} +{"id": "train_044574", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0a2ffd79e8d0a4a1dfa1a95d6731de6cf60eb2b7"} +{"id": "train_044575", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "e512cb7d3bbca18c722a2e5435d06f2cd5a583ac"} +{"id": "train_044576", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "2b5fce926d1d2978eeba691174fd42d3a40d4ad0"} +{"id": "train_044577", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "3fdce2327148066459b02a004bf8ce82ba9ad5b3"} +{"id": "train_044578", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6639da6417862da547e1a7e0c5f7f70b252371bb"} +{"id": "train_044579", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2a0a8dc4c9da431dfc9c07c1c3bfb1505e56176c"} +{"id": "train_044580", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7264d42e56becd04ef0d270a2ef8754720ea8ca8"} +{"id": "train_044581", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a833ef6c370ba2868014b667f04d7bda55acb56c"} +{"id": "train_044582", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "75c27a07e27150d70ae16a71824e54765f04e373"} +{"id": "train_044583", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c1bd7f9e7588d94a57249713d3335a5747364127"} +{"id": "train_044584", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["data", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "07f1cdd0a08c214d3b2ca123bd0764d6bd3dc3fd"} +{"id": "train_044585", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b20b29ee798b67f99e73dad44e4c4ce406ae9b47"} +{"id": "train_044586", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c10d1edcaefffe1b1e05b046018d0110d991d835"} +{"id": "train_044587", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5007962380b1f1c3a2352db97a37b52adbf9dfbf"} +{"id": "train_044588", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "2d17dfb27c39f75d5cc5e5c99034c5490f91b02e"} +{"id": "train_044589", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "0f6f260a37e67301165692a565735807eec1af8e"} +{"id": "train_044590", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "e88831ad00e54a5a8d6cf1ebcd48a6e1692af4e6"} +{"id": "train_044591", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "227d21d22265f4224751b872d405b7b5c8c6accc"} +{"id": "train_044592", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "4898cf3e38bcec89b1bb91297c92dee10dc46651"} +{"id": "train_044593", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6925d6167c8f805aa13bc9ce358f4436562d42ba"} +{"id": "train_044594", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "5849c746b270dc4f90f747696f5b2b6d33ceab21"} +{"id": "train_044595", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "5f13dfd7f541fb2617f765eb0d8aad66fc091e01"} +{"id": "train_044596", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "4c6dfe009a02d679faee03beba5662cca58f2e09"} +{"id": "train_044597", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7689f86a3df1d28f7237955a68208eeffefc0840"} +{"id": "train_044598", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "b08b39c4d64d6b987472ad280f4bfd84e0b39caf"} +{"id": "train_044599", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "967f1a7ae4581cbfb0629beeb5afca7bbbc4a496"} +{"id": "train_044600", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "2161c55f6f295700acbe45cb87d6a3a73a234ca2"} +{"id": "train_044601", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6e75f3863e6124c7fc597bba01b1025e39b4fba7"} +{"id": "train_044602", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "f49e894d9d3e1ee61801eb41328976040998de95"} +{"id": "train_044603", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "06415bdc4cdc81f8cc40119e7a9d747b087615f4"} +{"id": "train_044604", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "f4df679c7ba202de10ae072a05117b8d2f57cb8c"} +{"id": "train_044605", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "54678bded36fae309b4763248cff59fc78daf49a"} +{"id": "train_044606", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "fd2c7185d81b46f941d0e3ba8e3e774a7bbf6077"} +{"id": "train_044607", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "3d72623e91a7adc17cdb3786f46ee4dd165b09ad"} +{"id": "train_044608", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e6850e89a2cfdca98a009d2205bf583ceb6140ab"} +{"id": "train_044609", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "177448a18483a1b93b937dbb703ee22ac9a8362b"} +{"id": "train_044610", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "69c7194df8539d61b6c58fc77e248c537fdbe8d1"} +{"id": "train_044611", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["coding", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "672779d1c363f5a05e68211522d4143faeedb3ba"} +{"id": "train_044612", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["evaluation", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "199c8e540cc9c0712b2ec55fcac6cf35f2dba9f5"} +{"id": "train_044613", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "b67a5288dd8dbb69f55ecd68c790607b62ea1a5c"} +{"id": "train_044614", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c1c05ce98280cada21015e02c96f605100365c63"} +{"id": "train_044615", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "46e99d60a6d20bd254819bb3146f43cfe6e02962"} +{"id": "train_044616", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "26e50be306689d348d6dd4994d5b184d7a69ebbc"} +{"id": "train_044617", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "be08ed6bd0c1673c35cf476d7170cf6f5437229d"} +{"id": "train_044618", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c09e34ee3e56c21c4a66061200743b34e6cf3373"} +{"id": "train_044619", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "f4f5a021e7042ff4690c24ecdfb6deeb3640ec41"} +{"id": "train_044620", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "eb57516477ce8a75298a364ae0954bbfa226d984"} +{"id": "train_044621", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5a84afe22d2fa89949a8b32f7af77d74c56e5364"} +{"id": "train_044622", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "1806ebe8f8da2aade4a9780c111d4f81f6becaf2"} +{"id": "train_044623", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e777da60e3b1e6491d0494835b3deb605b266b12"} +{"id": "train_044624", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6441bbeeb081a0a904f147e0f35ef2f52a1572d8"} +{"id": "train_044625", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7f0a8ea07740a6f5496f42d253e37231f5e1e9aa"} +{"id": "train_044626", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "19b78e43209f924625c0c51850f649c4cf862760"} +{"id": "train_044627", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e27dda8422c74bd772a791a9bc55581e52e038c5"} +{"id": "train_044628", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "3862ddef5dc8ce80c36ed4d9122dc54c7a02ab1b"} +{"id": "train_044629", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "480127f520b23062eb230c69722e53b8ede4b219"} +{"id": "train_044630", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "625a381c8e6dda56526688461af7f6af739b056c"} +{"id": "train_044631", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "b35497be8962717733b73533d4fcfc1a90b9ab9e"} +{"id": "train_044632", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d6d6a6dff147d0c25108b2a5639a11fb6d16e970"} +{"id": "train_044633", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "0ebe5138ff9e100c4627555326c11661d3b5a63e"} +{"id": "train_044634", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1c42caaf209dce5b06da5fbd3130828bc777b8c1"} +{"id": "train_044635", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "d441ed2714cc59665c1cf4eb3b5fb0a5392bf646"} +{"id": "train_044636", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4060c38a4567aab89a6c90ead82341e5e0789138"} +{"id": "train_044637", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "633033fdfb76d8e061c685dc92c9e067f9e9c13a"} +{"id": "train_044638", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["research", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e72f29b2937b7eb2797676596f4000ef256e3b89"} +{"id": "train_044639", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "79d25f3b7d3d61d327eb8ba0d1adc0f77a2e8127"} +{"id": "train_044640", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "315f168898838d7eac203876a9e8df2e5e9fdd4f"} +{"id": "train_044641", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "94eaf3604b600e490bb20fc7f8cbc3d371f8673d"} +{"id": "train_044642", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "d3ae880988a39c1465f861cd753adb03dcf024d5"} +{"id": "train_044643", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "92a69375df9aa8c64348537491a576dcd36a36dd"} +{"id": "train_044644", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "a68716e00f6042edb83b1ec6c4e3065967e5f7d5"} +{"id": "train_044645", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "34b9fe614272c7fa26e0edda7356b37016f7f1f7"} +{"id": "train_044646", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "18fd609639609c81d564d0b4e6cdad7b948f36ce"} +{"id": "train_044647", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["agent", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "2e5b6ece1873d9500b5f4eb187a93a007a5f26d0"} +{"id": "train_044648", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "31c785b950452b5756a3fa529e13b8f638326d3f"} +{"id": "train_044649", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "5643b607e2566d4c01894e6779fe6817e40ffb65"} +{"id": "train_044650", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b16b2599ac3901d62b50c44b94bc174878a60f99"} +{"id": "train_044651", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "eb2503d55cbe812fbc5e304c4d427087dff9aede"} +{"id": "train_044652", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "745dfc0ad33d327e91b3f52d995ef0d3bbb30aeb"} +{"id": "train_044653", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "aa6945b135415e68613707ba51927a5cb6435c27"} +{"id": "train_044654", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "848bde2a4b5ba786b14d5e58fdadf3d5d585f9ca"} +{"id": "train_044655", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c095edccfd191e392c70ec4fe01f75825f2e594c"} +{"id": "train_044656", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ea68c944abb5e7959c68100589c608b87738a317"} +{"id": "train_044657", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e98fc399dc21d97403a11692f57bd07df775ff43"} +{"id": "train_044658", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "b192d5b5a708896a94ee77a470741e920b9414ef"} +{"id": "train_044659", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "d355335e83eadf5177952573e5a56bf81f7eecf2"} +{"id": "train_044660", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "fb1aa25488865ca3895c956ba856d98aa65394ca"} +{"id": "train_044661", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "38d52395a460afea6be2d2064b28791c68259aba"} +{"id": "train_044662", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "888420f6a746982494ff01503c5b86922bc038c8"} +{"id": "train_044663", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "fb015b3cd8bd8154da4991dcf1910689adca2ff7"} +{"id": "train_044664", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4f6d393913582394fc98f7996615e6b53ef45352"} +{"id": "train_044665", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "b709d796aa5a5e039b51a790adbf38178fcc484d"} +{"id": "train_044666", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "1c42caaf209dce5b06da5fbd3130828bc777b8c1"} +{"id": "train_044667", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4b5dbf01714a9bc4304ef8891704c3b432d619c3"} +{"id": "train_044668", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b8468d58a956a5877081f619e128725b2fd35ecd"} +{"id": "train_044669", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "29dff1131862c16217bc399a1c409387c46d8dbd"} +{"id": "train_044670", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "9cea79e9f61562274ac2f149343fdc819b68f770"} +{"id": "train_044671", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "519a6271730a0ca3edacf50b62a694cfb845e254"} +{"id": "train_044672", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a436573b7afec9f24e590671a24cfe15b48d362e"} +{"id": "train_044673", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "0a8142ba4e6e3ca2f1f74bd22ac3d03598835071"} +{"id": "train_044674", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "e67773b682d8e309c4b00edf7e3ca7be59db59d1"} +{"id": "train_044675", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a59696228bbf22f740dd4f550c85817fb4c5564d"} +{"id": "train_044676", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "d39ceaa4ca505cf43b29e68429b2e2d8c8619275"} +{"id": "train_044677", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "23fda4cb48c34e2a4b50524501627add28a8079f"} +{"id": "train_044678", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "348fe4e0e893334d269e3e4c9adf647b24814766"} +{"id": "train_044679", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "dd86698f68648042d505f6bb290b609a60045f3b"} +{"id": "train_044680", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "8dcc289ed62178dc15c142cb57611bcccfcff1d3"} +{"id": "train_044681", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "4cfd282fec47cdfd4cc65c5973aa73b23249c177"} +{"id": "train_044682", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "f8041983270fa0784c44e72aec7aacfd48263542"} +{"id": "train_044683", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d5e257e0a566b7ccf68ff9dc7789b8d5c6813769"} +{"id": "train_044684", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f9300bb8df88bbe9e24c655c5f08851b6527dc93"} +{"id": "train_044685", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d0e00ee8dc47ee16ed2a47111b4a2fb5c4479a73"} +{"id": "train_044686", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "0cc615f989e17d797951782cab4d936e926fe873"} +{"id": "train_044687", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "fd044e02593d76f397c98095accc0f7de391eaf9"} +{"id": "train_044688", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "91f874d9323fd932e4e8f9dfed0398cf6a993c8b"} +{"id": "train_044689", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "14f8f3d8d8de104a58bde9f6be1d97a2c99ca92f"} +{"id": "train_044690", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9ad955921e266d0bda0dd7024c351dc69cf6f066"} +{"id": "train_044691", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "research"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9e033604c6dd9d4fc235e9a18432c706f82c2463"} +{"id": "train_044692", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "709c2dda26a94b26a646b4c6482161aff933adde"} +{"id": "train_044693", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "3c681c0a5743a7257bc1fd951cbdc780b05c5f3b"} +{"id": "train_044694", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "defb025f431cc28790b8af28073a51dd6dcc59f4"} +{"id": "train_044695", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["research", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "037f2633ab413c80c98768df8f5287e90da8bc47"} +{"id": "train_044696", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "9129a57d4685522d829bde37b2ad02667827d2c4"} +{"id": "train_044697", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "483c8d681f946885d4d043e04420d3f058d67c0b"} +{"id": "train_044698", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "9264f53a3d748382d215a75d62f3e06b8dabb2af"} +{"id": "train_044699", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b78bf5f1155c0def430a5c5cb8b3295a213e2027"} +{"id": "train_044700", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a2fbbce6a4e7413d86337f57aec831f1c5bc80b1"} +{"id": "train_044701", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "1ebdbf509b733aa22349026fdeaba9d939298494"} +{"id": "train_044702", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "beec9102f1fe3dfb6f90a672715e912d1e35e4e9"} +{"id": "train_044703", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "21cbb493e427fcbcb12df962c3ae928c7f185529"} +{"id": "train_044704", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "bdc5826598fa896af658672efeea68e21de27938"} +{"id": "train_044705", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "20887b6397fa94bebb9029c157d1061aed65fe79"} +{"id": "train_044706", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "df3272bd28bcb1170bc6fccc15fda9f07b27cefb"} +{"id": "train_044707", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "ae6319b29c3d21b0d9e25048fded357f17e24b48"} +{"id": "train_044708", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "ab08ece87e295142f89b4aa1848c7e1ed84daeae"} +{"id": "train_044709", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "4bd149a2e50eb3f9943c2bb1c4e68146c258be99"} +{"id": "train_044710", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "SQL", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "17136fa30bc19088336708480c8f18d80583d3e1"} +{"id": "train_044711", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "dd127e934ba4a3d9a8895bebc0f07c59b61368a3"} +{"id": "train_044712", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "1aeca1fea86d07a2acfb1cd86b403ca0d946351a"} +{"id": "train_044713", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "SQL", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8fb0a33b3879fcc4d65c562a5f7a8890e85e7092"} +{"id": "train_044714", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a6f9446156a731b951e6bb37557bc6fe867799c1"} +{"id": "train_044715", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "06880ac1e401e382ef7af054699c57c988e0d0a5"} +{"id": "train_044716", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f7e5b275e6f89e0dacb069e93b4e6ae13815ecbb"} +{"id": "train_044717", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "ed06ce600ab5bc678e3180ed50926bd1270b4bfe"} +{"id": "train_044718", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "afe31f83904180ba0ba9fb8c0d2de91669fb15bf"} +{"id": "train_044719", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "a4dba4051426a23c7d1703ff924b8885bb454976"} +{"id": "train_044720", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["data", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "af0e8cf3d68bf0e4f0760e6937b27a3fa3fae4cf"} +{"id": "train_044721", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["research", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "681eb3fae676401c84592460bbe48034b84f6291"} +{"id": "train_044722", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "310d82f64b537f42426ff181fa5a4331d7f1434a"} +{"id": "train_044723", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "514863eb39f4a903709471a5ea12a8e5b4a315e9"} +{"id": "train_044724", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "5b947e81a2b2695d23ee5375b8d7ec7987791548"} +{"id": "train_044725", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "39fe795d006d9d96434fb6cb67f6bf3f7fbd3b96"} +{"id": "train_044726", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "47842924c69d1f34fa7a54c26efb8708c8711e3e"} +{"id": "train_044727", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "1ba3d0f41b6512141802fec8ba97ed0c70974d17"} +{"id": "train_044728", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "f4b5faadd65d698bb254856ea70ff0f2a372a47c"} +{"id": "train_044729", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "9c4f090499e344ef3e40cc4e57ec8eff221e2165"} +{"id": "train_044730", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "da0dcf3512fd0a64811a9da7e58368ecb7f7afcd"} +{"id": "train_044731", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "53a71df6609de6937664b5badc20d1990fc913d3"} +{"id": "train_044732", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d3e50a5cb047a5cd6f041dcb24095cb48ce68497"} +{"id": "train_044733", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "15e689c3d3d8672132739ded0e40130e1bb19d55"} +{"id": "train_044734", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["coding", "evaluation"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e823d97d25c01f1980221e1a6fc553bbfa30f86a"} +{"id": "train_044735", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3dd1ed2f298332daab5460758a0b2df6856ece18"} +{"id": "train_044736", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "322b7e4f8b8d86e9a15368c39bb260243a6ba0db"} +{"id": "train_044737", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "2a4b23553dc6a4d95482099da00b5fbd92f674bf"} +{"id": "train_044738", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "8151508714d125800d068dc7efad85c5147f0df6"} +{"id": "train_044739", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "e4278740ccc7e20a066f3b37c82b5bc4085826b1"} +{"id": "train_044740", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "09fd12951084bf22d1b6b61f608e3cc295a51e53"} +{"id": "train_044741", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9b6fb6b64c09c0753b43378c2643e8ef9e1c323b"} +{"id": "train_044742", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "07bbe8452edc0fdc09910306fdac99361fec2910"} +{"id": "train_044743", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "834ba8a8535cc230364f5f4678e38f1cb60cdce6"} +{"id": "train_044744", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "1bc2ca8386bd300096bba10f11a1f480ed24dfc1"} +{"id": "train_044745", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["coding", "data"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "54c809d44001c964aa65bac51393fd330a4dc43a"} +{"id": "train_044746", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "feabd844e83b5f3782050f116e6e75af946420ca"} +{"id": "train_044747", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4fb3ad483c7f6566d3273cfa715a3eb926ea6ea8"} +{"id": "train_044748", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "219d790455e9dd4be179bb87e4c9f2bbb0943e50"} +{"id": "train_044749", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "bdbae3857f1a3acea37413d7c5e758caebe855e9"} +{"id": "train_044750", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2c40bf61ee35834cc8c94a399da7390d2b8cc904"} +{"id": "train_044751", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "bf4117d0f42a525689c1ab5208720ce255d6a4ee"} +{"id": "train_044752", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "dbe9e3f1ab1f206d4bee86c3e6aae97d568a0342"} +{"id": "train_044753", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "8e0ff9c5e97003f04528f76457ee2eeed863ae38"} +{"id": "train_044754", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "5591b3e4b28a3a25e442b89784fab77af9786968"} +{"id": "train_044755", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "93237fccaf506dfb8346c68544fd5fdfa69d48a9"} +{"id": "train_044756", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "564e53b1b9fbe0fc2dff77934f3d6385b833bc5e"} +{"id": "train_044757", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["data", "evaluation"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "11c4278893cfac78edc4eca69ad21425c69be431"} +{"id": "train_044758", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "787dd58d6332b6ce8db1adda40d50c67f92e7fe4"} +{"id": "train_044759", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "b24a6cd1c9da599fa37c4f965aebab56e074298e"} +{"id": "train_044760", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "7fdd44fed47989e8c45b3c0e7259b378bdca27b3"} +{"id": "train_044761", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1aa86001e6c5c8310562148b3db2b5b396b92077"} +{"id": "train_044762", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6ac83ecaa6a787f054e5dc95a7c7a62b1a6f2dce"} +{"id": "train_044763", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a0a4588d1b30617f882e11011362493bb5212c5a"} +{"id": "train_044764", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "848a7e820ee13757aaedb2513935f56b1fe90478"} +{"id": "train_044765", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "fe7b6de905ffa2d36e8c28eaa95d8fe05a158b67"} +{"id": "train_044766", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "bbcdadc7981efb4208b571c559f4b89a9f12295a"} +{"id": "train_044767", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "805a1a45b48f461dcf81e7e6b44951b086a9d33f"} +{"id": "train_044768", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9f76a7db78be1cfbaf0c4e3cc7e5aedce0cb6507"} +{"id": "train_044769", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["security", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "8e7b13782149e8534cf5edf8eed541080630bb87"} +{"id": "train_044770", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["coding", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6cea7d758d2b223dbec508b61c056cb40023f5b6"} +{"id": "train_044771", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Rust", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d3fd87f4ffbad07a45e092f10d11c622c50c8b24"} +{"id": "train_044772", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "85ea21db941e346494ba415a17ac269dea538e15"} +{"id": "train_044773", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3739955e452c383df60ebb8879a0fbc904ae8500"} +{"id": "train_044774", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "400ed9801dd4fdf8bfb90082a6cf760027cb1421"} +{"id": "train_044775", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "353685efe6ce33f95582aaa8632f736248f0f932"} +{"id": "train_044776", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2c8bca35f6e2e5113e59040e3045432f921cc63b"} +{"id": "train_044777", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "7e93bc51aafbd97ff064305bc65870558919b5a0"} +{"id": "train_044778", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "9b1f8ac74635fbe22cd779bb2b52a3a0f470a111"} +{"id": "train_044779", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["performance", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "10047dd8634837cdb32784eedfd195a52d73dc32"} +{"id": "train_044780", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["data", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3e7e741b7cc7e5d12d11cf6e0deed2bcc2a74115"} +{"id": "train_044781", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "80ed28ac686b830bf21ab12cc9457168d5f80f20"} +{"id": "train_044782", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["security", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "801f015235026752b493e3c1960ae6cdefa40537"} +{"id": "train_044783", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "c0e62a1f086f507c00b0cfee4f96124e35c5cec9"} +{"id": "train_044784", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "48f8f5ad05ea6be84af85a466b23d972bbfef3cf"} +{"id": "train_044785", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "coding"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1db73beeefa7999e70b6c8fa6d36a5208ce7769f"} +{"id": "train_044786", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "26d660a6626471eecc052a5127f0f4b6c7e87f22"} +{"id": "train_044787", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "55bf7c85f3c5dd121167a0da1b92edbad0c7a92f"} +{"id": "train_044788", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "bcf0ea9a3317c869e35e58a8cb084e4787ae236c"} +{"id": "train_044789", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8e5e2c00a8ac5f70002355608bd14808e8f6edad"} +{"id": "train_044790", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "evaluation"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f471955f0fa3edb2ec8d21fa57618cd21f4ddb42"} +{"id": "train_044791", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "cce79a4443c15f6f38b8370c3d08fc98bef6e46b"} +{"id": "train_044792", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "57f0697dd57be146127536516c915466ff52a611"} +{"id": "train_044793", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "5c4327c7fc2a811a6ace27580e4df3f6a653fea1"} +{"id": "train_044794", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "75331a5dcea7fc4e77f480573e03912328181090"} +{"id": "train_044795", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: TypeScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "fe759de682f08b5508e4fd77bad0b3f87db26fa1"} +{"id": "train_044796", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "da931355a193eadb08ffcd78e00f21bfdb751905"} +{"id": "train_044797", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "af5a4d1dd3cffb61f026597dd91024b7f40f9395"} +{"id": "train_044798", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f9b3c8c948782f29d9d75e685fdfb3b4f50d64f5"} +{"id": "train_044799", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "0b0275abfc308b9df3f2dffb86f58e0a7492fcbb"} +{"id": "train_044800", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["research", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "89dbeda0b44286a48f5d20edb76ea6d33c896229"} +{"id": "train_044801", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "507fa7c523bca59eee7aeb7100730ede127c86ef"} +{"id": "train_044802", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "11c4278893cfac78edc4eca69ad21425c69be431"} +{"id": "train_044803", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8b1d06b13f795f77e8cd43e80d823e410d727d6b"} +{"id": "train_044804", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "74dfab6e55e80c24fa784de70a256617452535ca"} +{"id": "train_044805", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "10f85a2aebc8480559d989334d4129339de2996a"} +{"id": "train_044806", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "540ed2ca75f83d0a6abb7e0b6c9d747301eda6fc"} +{"id": "train_044807", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "dc2e1e9a71437c14b29274b6963e030578973f52"} +{"id": "train_044808", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "research"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "6d83d3cd80c3cd3b14a0571d5404f1148b34ffc7"} +{"id": "train_044809", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "d276faffe954c29108803d62d2ba57e6a1eda206"} +{"id": "train_044810", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3501f9ce735bf72aeb5904beae896a5f6fdd9abe"} +{"id": "train_044811", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["evaluation", "security"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "69b12a3766eac5ab54b019bd585b13c26e3bc9d7"} +{"id": "train_044812", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "53c01646f6a61c90c24d70b765d94ceadf889e24"} +{"id": "train_044813", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "f078cfcec33b3624d105cbd880d8f2f686198973"} +{"id": "train_044814", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "65d996ce87d918309595e6aeca3ce88a3750ca98"} +{"id": "train_044815", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "11500a44c573016cdedf9cdbdc2e4df5036d28ec"} +{"id": "train_044816", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "f93eaa471a7e4b0692b1830c206192b339ea08eb"} +{"id": "train_044817", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1a783fa1a21ca8e77f0e71c5cf9cdb5ee6bf2d5a"} +{"id": "train_044818", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "03f83a6aba0fc6f0d805ac31819f9de0f47f31e5"} +{"id": "train_044819", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "91b7c3c19404928142c5f4577ab1b3c714e2b1dd"} +{"id": "train_044820", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e4f1fd802a4e0dc383658887273a5af4da8f53ea"} +{"id": "train_044821", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3d408bb18f315ce30725f226f3103e3e7addd271"} +{"id": "train_044822", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["data", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6cda1ca0774c1e47512fc07801b6986f07852a44"} +{"id": "train_044823", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "e1a68061c7099650c579d7aeeebec38cbd811118"} +{"id": "train_044824", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "ffb4ed3f0bfdef16fec32ec3ddc3fa3d50ae441d"} +{"id": "train_044825", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["coding", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c41d9eb8b0cb86d2f713bd665372d8703ae43e9f"} +{"id": "train_044826", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4f23aa62fe6420adddae9f95c8531d68a738e0c2"} +{"id": "train_044827", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "1d09a8fbd3642d284643227756c67f16b3715c1e"} +{"id": "train_044828", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "da8cfb30ad82b3d5d51b2a71c6fcc2daf25d3d46"} +{"id": "train_044829", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "243a5fab2c1f8f51b82494d7b03c10cb82ed373e"} +{"id": "train_044830", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "e65fd3d20073bd05b85461345ec4a12c1505de8c"} +{"id": "train_044831", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "7dcd24449e0d9bab9eba01434257b09506ba419e"} +{"id": "train_044832", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "a7aa17952e04ce72a6eb9c76787080a2e3841e4e"} +{"id": "train_044833", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0d2e9005fe85971c145ea07364d46f3906ac748f"} +{"id": "train_044834", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "62949c7c15652a54d636060dab82da713458fc84"} +{"id": "train_044835", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "450625fef5b2871c0fc6d029938afafce4f323cf"} +{"id": "train_044836", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "8706b13edfc8649878b6d321b812c6df0873753c"} +{"id": "train_044837", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1fd979b03800940acd0255f84aaab853b24f8513"} +{"id": "train_044838", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "e835241685c4476491613698d819e3db804edce8"} +{"id": "train_044839", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "ae26be2023d8fd4f7013171a764dcf14d0321cdb"} +{"id": "train_044840", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["agent", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2b6d4d820d32022f1f9d26751c4e8e1c6ba32294"} +{"id": "train_044841", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4bd7768e74ff1361a9cee4624f9a7d461fe39193"} +{"id": "train_044842", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "303bd770ea1e0eff94d22008fbc02cb58701362d"} +{"id": "train_044843", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "05c85bc47d1c59df89780f6766a62e2d08e1abce"} +{"id": "train_044844", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "ab1a9b6385e56ce1eb89d9d55cb87da3c9595ed7"} +{"id": "train_044845", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "SQL", "moe_experts": ["data", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1500640766415c5c43ff74dbfd245cf4602dfd19"} +{"id": "train_044846", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "044e805207dfbbde2625bbc39aaf4a8101562812"} +{"id": "train_044847", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["performance", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "3cffa656bbc8560ad00b210a6f8f8a5a87ac472b"} +{"id": "train_044848", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "3c06c48f69852f320906c5cd2155cdc09d98ae9d"} +{"id": "train_044849", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Go", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "72ccf14c51952945d265ab469d318aa926b44b80"} +{"id": "train_044850", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "47d3da19ce702c5c99552914e2ee51ca620f77a9"} +{"id": "train_044851", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Python", "moe_experts": ["agent", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "397fb668c8e1df0ab36a92fab80544325d1f7b84"} +{"id": "train_044852", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["security", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "36b13dbae2b0f51a5aff2cb494a16635d1a2490c"} +{"id": "train_044853", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "334af81e571b2058d9cf8463827272ba49ffaa37"} +{"id": "train_044854", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "c08c3d308c8588aa781101a1e1ec35227c49f880"} +{"id": "train_044855", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "coding"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "9fdb1f4c1ba0a2bddcbdcce68899bd1208c6dd8d"} +{"id": "train_044856", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["security", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "10e33259d467a45c9470fc4f1d44c3ca6fc35463"} +{"id": "train_044857", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "61d8e4f042e6b467bb6249a155d9d3017919e6cf"} +{"id": "train_044858", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "f3d28b77666c89dec6086c8992b241be715fc285"} +{"id": "train_044859", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "41a12146278e1e605b416abb60957dcbcb7c85e5"} +{"id": "train_044860", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "2246bd32d11285164802ae185b9d75615c1fdcda"} +{"id": "train_044861", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Go", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "e09ebe4f94c453d3e2b02493a9a7c3cd18df8aa5"} +{"id": "train_044862", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "8d39d2183ccaa901177b480eeda37a4e36809a44"} +{"id": "train_044863", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "741559f10f38dcc96e18bd25336f0724357fea80"} +{"id": "train_044864", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "57f0697dd57be146127536516c915466ff52a611"} +{"id": "train_044865", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["security", "research"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "25bb33755b010cb03f07fcd06230a37ebb60fb22"} +{"id": "train_044866", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "848a7e820ee13757aaedb2513935f56b1fe90478"} +{"id": "train_044867", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "50fdb92db0b7fc2ce79a40bae8d665fb7321ccef"} +{"id": "train_044868", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: SQL\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3eee0f5ce6492e935f9fe65a97ec54047d27445d"} +{"id": "train_044869", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "2352a90e8da764a302f66c467a938ef51362f92b"} +{"id": "train_044870", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c5d75035739fc009f291f4ecfa77984185243783"} +{"id": "train_044871", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Python", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "598855efd1b19f947df9628d0820751b68438e57"} +{"id": "train_044872", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "64a655ffbf298d843456b63f7bf1c3216d057de9"} +{"id": "train_044873", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4c9983803552b6880411af09501d5a22bd11e351"} +{"id": "train_044874", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["evaluation", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "a7d6c29d04649f519965f86d07dea746490fd77f"} +{"id": "train_044875", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["coding", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "78b99fe496d23c35d88aeee3e1bc28dcb7001194"} +{"id": "train_044876", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["agent", "performance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "18370f25df943f87a424945545ec1505e3d989d9"} +{"id": "train_044877", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "bd46bc4f68fb995eaec63cb832307b236b01485d"} +{"id": "train_044878", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "00176edef004ffa2eb91253efd9adde9fa7893f2"} +{"id": "train_044879", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "7434cd007340c5c18fa6e18e63c42e26f7864bf6"} +{"id": "train_044880", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "b7d5770249c22d350198d4a0b29642e71bbfe21e"} +{"id": "train_044881", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["evaluation", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "f2811b87bed84fff6caac4f6b4e58faac598ef4c"} +{"id": "train_044882", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "945c3a3c3a6c28cb1ed367dce93901022bebff59"} +{"id": "train_044883", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "99ad28a8c1930642aca742d9e53ec08fb8b6145a"} +{"id": "train_044884", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Go\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "6f0bc8c2befdced3d8ba543e7cac06c88ae19877"} +{"id": "train_044885", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "b269a5c3efc95e97ccb22a1d81bfc44ba426da52"} +{"id": "train_044886", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "27570ceee3d92ec0a45843810bf051742ec78c99"} +{"id": "train_044887", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "evaluation"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "750f9e979391511c54072335c6b351baf77d1412"} +{"id": "train_044888", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "4a5f8ef2fca214705fd79c18ad39ba618c36f9d1"} +{"id": "train_044889", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Java", "moe_experts": ["evaluation", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "de22ddf12fbbbc0ea1f66f0b5d2038ecf3d53583"} +{"id": "train_044890", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "735ea102076d4ebdcc2bf1ca1ff3b3fcad6fed2f"} +{"id": "train_044891", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Bash", "moe_experts": ["security", "coding"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "993843a922ccf6152344df669ec67423e3c7db8f"} +{"id": "train_044892", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "cb6ff96f470954817af932b7b512c13e48740be0"} +{"id": "train_044893", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["evaluation", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "bd280ae21ea53567a42c61c5d1aac1cad1518a57"} +{"id": "train_044894", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "667e77a831395214e98cdeb027219ccdafc9184c"} +{"id": "train_044895", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["research", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "88a480b363c62279c91530e52f477c493d9d66b1"} +{"id": "train_044896", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "b02ecc8c5f98dbed08892fa479ca5544fd007f7c"} +{"id": "train_044897", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "308b2a5a571e09ba297df63992a31f80c7b7d0a7"} +{"id": "train_044898", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: TypeScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "TypeScript", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "58aaf0d8932d94daac16196fcba768ac88d60747"} +{"id": "train_044899", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["research", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "5efd934988ddfa4bba49211422e51016eeb43c4e"} +{"id": "train_044900", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "SQL", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "da346e94ab19c3aa953f96388a30d76fd5eee4d6"} +{"id": "train_044901", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "evaluation"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "c14ab374b8584a723ee18b082e52b7a6cdb91a2d"} +{"id": "train_044902", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "advanced", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8f6edb883b07d9bf6b9153240a14a3f8dab4a699"} +{"id": "train_044903", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["evaluation", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "29cbd4deae69394b34eea2d25c010256aad71d75"} +{"id": "train_044904", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["security", "governance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "048d6ba716a3fa70fce2c4b47e280c47c87f7243"} +{"id": "train_044905", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "13ff421d450fad48a9b0c633c75317c56ab9e73d"} +{"id": "train_044906", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["coding", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "05a2950bae1ee0126a35f406d2914325e60a5727"} +{"id": "train_044907", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d5ae096e703c4681511e872f7303b3e915e70740"} +{"id": "train_044908", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["data", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "4c048aa84de53b6e1df9d1794b89347bbb01e0a8"} +{"id": "train_044909", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "TypeScript", "moe_experts": ["data", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "81fc2ca908ea801fdc2abb562ff29f81e2987c6d"} +{"id": "train_044910", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "ff502b1f0124dcd39a42ec9ab431fff7b513cf22"} +{"id": "train_044911", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: Bash\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Bash", "moe_experts": ["data", "performance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "90e64ebec8e27849dc0876854f2d87827899bd88"} +{"id": "train_044912", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["evaluation", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "b5911f1221a78f93d3f72a38150082e9b79ead07"} +{"id": "train_044913", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["evaluation", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "8c84edbdbda008d1623dccba98a505d75db23629"} +{"id": "train_044914", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Java\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0c97ca25407e5026c6d751746b18231a3e79f41d"} +{"id": "train_044915", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "1fe9980642aebe7e60a535c7dc0d86f8ff6cee8d"} +{"id": "train_044916", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["coding", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "645013386f862c2ae542f049cdcf94d7d0655a26"} +{"id": "train_044917", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["performance", "data"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "87d32bb980db05d82aa8bb9f76b5d1da78ecf36c"} +{"id": "train_044918", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: SQL\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "agent"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "3b64cf98d90bacefca48ae9c7b4cd9ee1cbdcf21"} +{"id": "train_044919", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "df16832835a5b96adfc5a55e716d3f2e9effbe9e"} +{"id": "train_044920", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "3f6da69d11c9d427ec30a77c14fb2d554ccca49b"} +{"id": "train_044921", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["security", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "9c2d1fb224630cea2007058188358defde82ec02"} +{"id": "train_044922", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "eval", "difficulty": "advanced", "instruction": "Task: eval\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "C#", "moe_experts": ["governance", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "e6e4af2472d05beeb792804f810baa4a22133f17"} +{"id": "train_044923", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "39ade47c9ab33873eb8b433c357df15e748a7e68"} +{"id": "train_044924", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["evaluation", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0ba1a00d051ecbf2c214d045800768528f5dc729"} +{"id": "train_044925", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "2a3dcab15a9f8184d906fddee28addc0f268219f"} +{"id": "train_044926", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "data_pipeline", "difficulty": "advanced", "instruction": "Task: data_pipeline\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "4427c76040c63ae5859e1442f0b9fe3f2c1012b9"} +{"id": "train_044927", "created": "2026-01-01", "topic": "Multi-agent coordination and role separation", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Multi-agent coordination and role separation\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["data", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "cf280f76df1acc6cd07f95e70c623b0cf697fb80"} +{"id": "train_044928", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "d4cc4fd29f8ab7dba7b501d978529e0551836759"} +{"id": "train_044929", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Python\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["evaluation", "data"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "c3daa895800341a04625ad94844e211d9b88ac84"} +{"id": "train_044930", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["agent", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "48e923d4e1d9977a3aa7eb540aa57e07f007abc8"} +{"id": "train_044931", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Go", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "4e59b2de9c5074c4fe7269f071b5fa448fb2948f"} +{"id": "train_044932", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Go\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["coding", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "d4fb839a59fd1bcf21592b6ab1660b48498643dc"} +{"id": "train_044933", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "data"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a46a18e22f6ccf0abbd595908198b06be92c6aa3"} +{"id": "train_044934", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Bash\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["research", "governance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "81e2b0951190ec9096db73fab973918e701547ea"} +{"id": "train_044935", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "fdb9bc84cd73785e32952b514cd554740abc9cc2"} +{"id": "train_044936", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["governance", "performance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "1e056b810dce42cd27b4aa64380307ec9ff2960d"} +{"id": "train_044937", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d6b2d5cdcfd44c0adcded8916c59389254b4f409"} +{"id": "train_044938", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["coding", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "d8cbd5a35bbf0f831b928ba3459654772561608d"} +{"id": "train_044939", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["evaluation", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "8eb7697a47afd91657cb1b85a8730ee07c32cd10"} +{"id": "train_044940", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: expert\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 9,\n \"confidence\": 0.9,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Go", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "2065d8433ef1c3d34ea1781f67056121ff317086"} +{"id": "train_044941", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "eval", "difficulty": "expert", "instruction": "Task: eval\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nMetrics:\n- pass@k\n- time-to-green\n- cost-per-fix\n- regression rate\n", "metadata": {"language": "Rust", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "5847ac4d08a03e24fc826788d98c81c66d2b8dd1"} +{"id": "train_044942", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "compare", "difficulty": "expert", "instruction": "Task: compare\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["security", "research"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "a90edc01b3a478134e88f413dd8ab7c987a41883"} +{"id": "train_044943", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["coding", "performance"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "680eb2e6b1042209d16fde07a52c8cb46c45801f"} +{"id": "train_044944", "created": "2026-01-01", "topic": "Reasoning-first coding models and tunable deliberation", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Reasoning-first coding models and tunable deliberation\nDifficulty: advanced\nLanguage: C#\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "C#", "moe_experts": ["performance", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "dce93b042cfa2518401973623e98b1080781fc8b"} +{"id": "train_044945", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "governance"], "cost_budget": 4, "latency_ms_target": 50, "audit_required": true}, "hash": "a8999b23e59d4b6fd3851b2ce22c126ae3a8c0d0"} +{"id": "train_044946", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["agent", "governance"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "4ae32b39e316f24b98bd27f47d406b348ed9ff4a"} +{"id": "train_044947", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "C#", "moe_experts": ["data", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "a5f8df7329d4ff36e7f99c79684ea3ecfc6400cb"} +{"id": "train_044948", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "explain", "difficulty": "expert", "instruction": "Task: explain\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9edffdab2dfe17bdea443c6b673ad105cb23ce6f"} +{"id": "train_044949", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "security"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "82632e59b5b5891f9fd38ab9f241b6d928fc8906"} +{"id": "train_044950", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Rust\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["governance", "research"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "0da11b72f051a1906a1a67597f409db8fa042a61"} +{"id": "train_044951", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["security", "governance"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "fe4b43e4036917292ce82f1b2d34ddcb77b3384b"} +{"id": "train_044952", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: Python\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["performance", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "9c7703d27857930a8ee4b67e42423e7cc8978395"} +{"id": "train_044953", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: JavaScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "205e407254a8a5975357f089d04b90e6156d9ff1"} +{"id": "train_044954", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: C#\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["evaluation", "coding"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "8ccd248c9164d5861082df2e1a4017a7ee3a089f"} +{"id": "train_044955", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Governance, provenance, and audit logging\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "agent"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "6ea0421351f2754f6dfe0940da1c20887cca591c"} +{"id": "train_044956", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "governance"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "ee47fb252944a86e6fac0ffcb441b4ae2027cffc"} +{"id": "train_044957", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "42463af8a7e73857de1b3d4e68b132ce25808573"} +{"id": "train_044958", "created": "2026-01-01", "topic": "Dataset curation pipelines and quality attribution", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Dataset curation pipelines and quality attribution\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["research", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "646735f7dc0db4ac01ea3cc12b9c2b25a9a1b4f1"} +{"id": "train_044959", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "645f72029fa5e175ba1d6580a53a64084ff468e7"} +{"id": "train_044960", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["evaluation", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "aeed446e307f7f446fd6ef290ae2badbe0b1e12f"} +{"id": "train_044961", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "evaluation"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a15c2c23d33decfc9186ac08e8d3478fe50c463b"} +{"id": "train_044962", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "JavaScript", "moe_experts": ["governance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8c07a5a07063b6314fe215fa782c331e26cc2c92"} +{"id": "train_044963", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 6,\n \"confidence\": 0.6,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Python", "moe_experts": ["data", "evaluation"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "e6832acdd52f93b03c696a6edb65daa3a3d2fbd5"} +{"id": "train_044964", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "evaluation"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9e750cf447fae0335f50ede631a1002c5697b215"} +{"id": "train_044965", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Bash\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Bash", "moe_experts": ["data", "research"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "fa0dbd5f7c8cd011afa50a44e0bfa9411189dd25"} +{"id": "train_044966", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["agent", "security"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "6538a4711c511898af65193f06810b2dc426ff22"} +{"id": "train_044967", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: SQL\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["research", "governance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "d53c1afab175e8cd2822d171d74fb9e6eb4ad96f"} +{"id": "train_044968", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["agent", "performance"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "48d60c823df17306c1ccebe12c38b6c1bc03e1ad"} +{"id": "train_044969", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "failure_analysis", "difficulty": "expert", "instruction": "Task: failure_analysis\nTopic: Code-specialized model families and scaling laws\nDifficulty: expert\nLanguage: Rust\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nFailure: regression detected\nCause: missing edge case\nFix: add guard + test\n", "metadata": {"language": "Rust", "moe_experts": ["performance", "agent"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "37d3cd971077247bcd428062ae1ffe53b8d8992d"} +{"id": "train_044970", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Latency, cost, throughput optimization\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 10,\n \"confidence\": 1.0,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Java", "moe_experts": ["data", "governance"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "6a0810384e619686bb7582c5ed998e2f8f6e4147"} +{"id": "train_044971", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Python", "moe_experts": ["performance", "coding"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "aa0f11aea096c6d88c443eba07bba6cdd3db6d50"} +{"id": "train_044972", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["agent", "coding"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "dc65380ef35a5c2abe43ecb4c8781934a28d2529"} +{"id": "train_044973", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: SQL\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "SQL", "moe_experts": ["performance", "coding"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "8a775ba41acc62210734cbdff687cd298efdc467"} +{"id": "train_044974", "created": "2026-01-01", "topic": "SWE-bench real-repo evaluation and leakage control", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: SWE-bench real-repo evaluation and leakage control\nDifficulty: expert\nLanguage: JavaScript\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["research", "data"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "5908ec8c26b52c257898650cf5402e7154a34c05"} +{"id": "train_044975", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Bash", "moe_experts": ["security", "evaluation"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "09e366834423600bcc06e29004d2299a3b5d0d2f"} +{"id": "train_044976", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "code", "difficulty": "expert", "instruction": "Task: code\nTopic: Failure engineering and resilience testing\nDifficulty: expert\nLanguage: Python\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["research", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "803c2f912daf45686171e13ae54d45e30676c42c"} +{"id": "train_044977", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "review", "difficulty": "expert", "instruction": "Task: review\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: C#\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["governance", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "c55049e7ba8f3ebc943f029ca7c26301e21af6a3"} +{"id": "train_044978", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: Rust\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "a5a900a3f30a702d5c7edb210e58bd2fe1a9dc9d"} +{"id": "train_044979", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "agent_loop", "difficulty": "advanced", "instruction": "Task: agent_loop\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: advanced\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "data"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "c321abf91a6404a994ab2fe42fea3943c4b57b35"} +{"id": "train_044980", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Rust", "moe_experts": ["performance", "research"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "643839cde00ebdcdbff3fd28e55c52931562d6f0"} +{"id": "train_044981", "created": "2026-01-01", "topic": "Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Agentic coding systems (plan\u2192edit\u2192test\u2192reflect)\nDifficulty: expert\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["security", "agent"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "c47ef460361f44678a3dccac212b348ff2f09daa"} +{"id": "train_044982", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "agent_loop", "difficulty": "expert", "instruction": "Task: agent_loop\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: expert\nLanguage: TypeScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "2490dbef629598c6732995e30a1e1f48f2b25bca"} +{"id": "train_044983", "created": "2026-01-01", "topic": "Secure code generation, red-teaming, and policy gates", "task_type": "self_grade", "difficulty": "expert", "instruction": "Task: self_grade\nTopic: Secure code generation, red-teaming, and policy gates\nDifficulty: expert\nLanguage: Bash\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 7,\n \"confidence\": 0.7,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "Bash", "moe_experts": ["agent", "security"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "334542ac7b676e13c5401b709255b8dfac4d935a"} +{"id": "train_044984", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Extended context, memory, and repo graphs\nDifficulty: advanced\nLanguage: Python\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nCode:\n```python\ndef agent_loop(plan, edit, test, issue, budget=4):\n cost = 0\n for _ in range(budget):\n p = plan(issue)\n patch = edit(issue, p)\n ok, _ = test(patch)\n cost += 1\n if ok:\n return patch, cost\n raise RuntimeError(\"Budget exhausted\")\n\n```", "metadata": {"language": "Python", "moe_experts": ["agent", "security"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "20cb57e2efd0c273d88e3f63c6da48b30f4933a1"} +{"id": "train_044985", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "code", "difficulty": "advanced", "instruction": "Task: code\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "C#", "moe_experts": ["research", "performance"], "cost_budget": 2, "latency_ms_target": 200, "audit_required": true}, "hash": "9e769f1b4dd5af23f351e55c923dca9c8b53673a"} +{"id": "train_044986", "created": "2026-01-01", "topic": "Extended context, memory, and repo graphs", "task_type": "design", "difficulty": "expert", "instruction": "Task: design\nTopic: Extended context, memory, and repo graphs\nDifficulty: expert\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["data", "performance"], "cost_budget": 5, "latency_ms_target": 200, "audit_required": true}, "hash": "c54d7100666aead0ed84223da9bd0d548bd4e68d"} +{"id": "train_044987", "created": "2026-01-01", "topic": "Tool calling protocols and sandboxed execution", "task_type": "patch_diff", "difficulty": "advanced", "instruction": "Task: patch_diff\nTopic: Tool calling protocols and sandboxed execution\nDifficulty: advanced\nLanguage: JavaScript\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "JavaScript", "moe_experts": ["performance", "coding"], "cost_budget": 2, "latency_ms_target": 100, "audit_required": true}, "hash": "05c85bc47d1c59df89780f6766a62e2d08e1abce"} +{"id": "train_044988", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "self_grade", "difficulty": "advanced", "instruction": "Task: self_grade\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: C#\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n{\n \"score\": 8,\n \"confidence\": 0.8,\n \"notes\": \"Passed tests; minimal diff; no security flags.\"\n}", "metadata": {"language": "C#", "moe_experts": ["evaluation", "research"], "cost_budget": 4, "latency_ms_target": 100, "audit_required": true}, "hash": "471ed447bc97b72f494ff2eec2c880948ce38f81"} +{"id": "train_044989", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Model merging, distillation, and continual learning\nDifficulty: advanced\nLanguage: Java\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["governance", "data"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "29dff1131862c16217bc399a1c409387c46d8dbd"} +{"id": "train_044990", "created": "2026-01-01", "topic": "Failure engineering and resilience testing", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Failure engineering and resilience testing\nDifficulty: advanced\nLanguage: Go\nContext: Latency-critical production API\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["evaluation", "performance"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "0d54c1976e15efa9750d83cea0bdaeb96afe01ce"} +{"id": "train_044991", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "tool_trace", "difficulty": "advanced", "instruction": "Task: tool_trace\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: Java\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "Java", "moe_experts": ["agent", "security"], "cost_budget": 5, "latency_ms_target": 100, "audit_required": true}, "hash": "3a0312cf2866faf7b6b6fe17bbd9e9d38b95fba1"} +{"id": "train_044992", "created": "2026-01-01", "topic": "Code-specialized model families and scaling laws", "task_type": "compare", "difficulty": "advanced", "instruction": "Task: compare\nTopic: Code-specialized model families and scaling laws\nDifficulty: advanced\nLanguage: Go\nContext: Research benchmark validation environment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Go", "moe_experts": ["performance", "security"], "cost_budget": 4, "latency_ms_target": 200, "audit_required": true}, "hash": "76c9892c2bd8e61f0838cae655deb0d9803afe1f"} +{"id": "train_044993", "created": "2026-01-01", "topic": "Latency, cost, throughput optimization", "task_type": "design", "difficulty": "advanced", "instruction": "Task: design\nTopic: Latency, cost, throughput optimization\nDifficulty: advanced\nLanguage: SQL\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "SQL", "moe_experts": ["governance", "coding"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "6a3241fc19a18814736c053d2fee58be00db08ef"} +{"id": "train_044994", "created": "2026-01-01", "topic": "Governance, provenance, and audit logging", "task_type": "patch_diff", "difficulty": "expert", "instruction": "Task: patch_diff\nTopic: Governance, provenance, and audit logging\nDifficulty: expert\nLanguage: Python\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nPatch:\n```diff\n- result = a / b\n+ if b == 0:\n+ raise ValueError('division by zero')\n+ result = a / b\n\n```", "metadata": {"language": "Python", "moe_experts": ["governance", "agent"], "cost_budget": 3, "latency_ms_target": 50, "audit_required": true}, "hash": "5572610aa236981ee1a59a2f31bea94fa444aa4a"} +{"id": "train_044995", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: Rust\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Rust", "moe_experts": ["research", "data"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "3bad62da9c3792a40648b0e1f7158f392e1f388e"} +{"id": "train_044996", "created": "2026-01-01", "topic": "Model merging, distillation, and continual learning", "task_type": "tool_trace", "difficulty": "expert", "instruction": "Task: tool_trace\nTopic: Model merging, distillation, and continual learning\nDifficulty: expert\nLanguage: C#\nContext: Regulated enterprise with audit requirements\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\n[\n {\n \"tool\": \"repo_checkout\",\n \"args\": {\n \"commit\": \"abc123\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"unit\"\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"files\": 1\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"full\"\n }\n }\n]", "metadata": {"language": "C#", "moe_experts": ["security", "agent"], "cost_budget": 5, "latency_ms_target": 50, "audit_required": true}, "hash": "240462aaf8fe71b56565a10386e39eea42511e19"} +{"id": "train_044997", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "explain", "difficulty": "advanced", "instruction": "Task: explain\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: advanced\nLanguage: TypeScript\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "TypeScript", "moe_experts": ["data", "security"], "cost_budget": 2, "latency_ms_target": 50, "audit_required": true}, "hash": "8d3116243664cd204c12ad42ad36a513203440e1"} +{"id": "train_044998", "created": "2026-01-01", "topic": "Self-grading and automated model critique", "task_type": "review", "difficulty": "advanced", "instruction": "Task: review\nTopic: Self-grading and automated model critique\nDifficulty: advanced\nLanguage: Java\nContext: Large monorepo with flaky tests and strict CI\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "Java", "moe_experts": ["coding", "security"], "cost_budget": 3, "latency_ms_target": 200, "audit_required": true}, "hash": "36acfbed3e33a6b57208726a795a8135d4f4821b"} +{"id": "train_044999", "created": "2026-01-01", "topic": "Mixture-of-Experts (MoE) routing and sparsity", "task_type": "data_pipeline", "difficulty": "expert", "instruction": "Task: data_pipeline\nTopic: Mixture-of-Experts (MoE) routing and sparsity\nDifficulty: expert\nLanguage: JavaScript\nContext: Offline air-gapped deployment\nProduce frontier-grade, production-valid output.", "input": "", "output": "Facts:\n- Modern AI coding success is measured by test pass rates, iteration cost, and reliability.\n- Agentic and multi-agent systems outperform single-pass generation.\n\nExpert guidance with risks, metrics, and governance controls.", "metadata": {"language": "JavaScript", "moe_experts": ["data", "research"], "cost_budget": 3, "latency_ms_target": 100, "audit_required": true}, "hash": "1bbd2f612877730ccfc4a382ad5a693bad634ab0"}