| {"id": "p_0119db8051", "text": "Add that rule to the CLAUDE.md so you stop asking me about it.", "target_terms": ["CLAUDE.md"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_03cf941609", "text": "We're testing a small language model that fits on a laptop.", "target_terms": ["small language model"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_07b51b33f2", "text": "The match should respect word boundary rules, so substrings never count as hits.", "target_terms": ["word boundary"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_083ced4bc9", "text": "The code review found two bugs, fix those and leave the style comments.", "target_terms": ["code review"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_085d18bd1d", "text": "Summarise the diff in two sentences for the commit message.", "target_terms": ["diff"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_089de36853", "text": "Can you rewind the video to the part where he explains the config?", "target_terms": ["rewind"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_08aff22141", "text": "A small language model is plenty for routing and classification.", "target_terms": ["small language model"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_08e4b0fe21", "text": "Which inference provider gives the best price for the open models?", "target_terms": ["inference provider"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_0a60c321ba", "text": "Does CLAUDE.md still mention the old branch name, or did we fix it?", "target_terms": ["CLAUDE.md"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_11c5307373", "text": "Did you read the CLAUDE.md before you started, because it says not to run local builds.", "target_terms": ["CLAUDE.md"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_1325a2c4d0", "text": "We filled the gaps with synthetic data, but the style was a bit too clean.", "target_terms": ["synthetic data"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_137d05fa15", "text": "I wrote the tool schema by hand, so I should check it against the code.", "target_terms": ["tool schema"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_1950398867", "text": "Do a code review of the diff before I merge it.", "target_terms": ["code review"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_1e12a7379c", "text": "Does the Agent SDK load the project's CLAUDE.md by default?", "target_terms": ["Agent SDK"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_24b0a58038", "text": "The status line stopped updating after the last change, check the script.", "target_terms": ["status line"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_273325c96a", "text": "Run that subagent on Fable and compare its answer with yours.", "target_terms": ["Fable"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_311ef3827c", "text": "Move the environment notes out of CLAUDE.md and into a separate context file.", "target_terms": ["CLAUDE.md"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": "fixed: Near-duplicate of existing bank sentence (\"Add that rule to the CLAUDE.md so you stop asking me about it.\").", "term_origins": ["claude"]} |
| {"id": "p_32cd6213fb", "text": "Wrap the regex in a word boundary check so short acronyms do not match inside words.", "target_terms": ["word boundary"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": "fixed: Stiff passive textbook definition copying pack meaning and redundant with rows 156-157.", "term_origins": ["daniel"]} |
| {"id": "p_37bea14bc9", "text": "What's the status of the PR from this morning?", "target_terms": ["PR"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_4107b5a9db", "text": "Is Fable available in the model picker on my plan?", "target_terms": ["Fable"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_4266b7c339", "text": "Add the current branch and the model name to the status line.", "target_terms": ["status line"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_46942e62c7", "text": "Read CLAUDE.md at the start of the session before you touch any files.", "target_terms": ["CLAUDE.md"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": "fixed: Textbook definition copying pack meaning; rewritten as a natural agent instruction.", "term_origins": ["claude"]} |
| {"id": "p_4841670d59", "text": "I pasted the old cursorrules back in after the rewrite, so they are duplicated.", "target_terms": ["cursorrules"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_48b6a35a8c", "text": "Lower the reasoning effort for the bulk renames, they don't need deep thought.", "target_terms": ["reasoning effort"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_4e0ccc73f0", "text": "Keep cursorrules short so Cursor does not drop half of it.", "target_terms": ["cursorrules"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_509948e4cf", "text": "Never let the validation set overlap with the training recordings.", "target_terms": ["validation set"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_53100b4242", "text": "Pass a lower reasoning effort on the classification pass so the batch finishes before lunch.", "target_terms": ["reasoning effort"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": "fixed: Near-duplicate of existing bank sentence (\"Lower the reasoning effort for the bulk renames, they don't need deep thought.\").", "term_origins": ["claude"]} |
| {"id": "p_534d458b71", "text": "Turn the reasoning effort up to high for this one, it's a tricky bug.", "target_terms": ["reasoning effort"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_5b7476e0a3", "text": "For this feature, a small language model will cut our latency and our bill.", "target_terms": ["small language model"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_5bd7841848", "text": "The kids spent the whole afternoon in the sandbox at the park.", "target_terms": ["sandbox"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_5bf0498197", "text": "Check the Agent SDK docs for how subagents are defined in code.", "target_terms": ["Agent SDK"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_5dd441604c", "text": "The validation set loss stopped improving after the third epoch, so we stop there.", "target_terms": ["validation set"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_60ee8cf391", "text": "Make the term matcher stop at each word boundary, not inside longer words.", "target_terms": ["word boundary"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_6298b1a65d", "text": "The results improved once we mixed synthetic data with the real recordings.", "target_terms": ["synthetic data"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_63ddbf036f", "text": "Open a PR for this and merge it, don't leave it waiting for me.", "target_terms": ["PR"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_64bd421109", "text": "Keep reasoning effort on medium by default and only bump it for hard bugs.", "target_terms": ["reasoning effort"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": "fixed: Unidiomatic phrasing (\"Is the reasoning effort worth it\" without specifying high effort).", "term_origins": ["claude"]} |
| {"id": "p_71b9a6d59d", "text": "The sandbox is blocking network access, so that's why the download failed.", "target_terms": ["sandbox"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_758abaa7e9", "text": "Explain how our retrieval-augmented generation pipeline picks which passages go into the prompt.", "target_terms": ["retrieval-augmented generation"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_767bf91931", "text": "Our inference provider had an outage, so the whole pipeline stalled for an hour.", "target_terms": ["inference provider"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_780938cacc", "text": "Please rewind to before the last edit, that change broke the service worker.", "target_terms": ["rewind"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_7893d16222", "text": "Is the browser a deferred tool in this session or already loaded?", "target_terms": ["deferred tool"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_7a8dc11c03", "text": "Does adding synthetic data from a stronger model count as distilling it?", "target_terms": ["synthetic data"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_7cbc7e632a", "text": "The PR has conflicts with master, fix them and push again.", "target_terms": ["PR"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_7cf6eeba05", "text": "Ask Fable for a children's fable about a tortoise and a hare who both want the last biscuit.", "target_terms": ["Fable"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_8d6f593964", "text": "With retrieval-augmented generation, bad chunking fails quietly, so inspect the chunks first.", "target_terms": ["retrieval-augmented generation"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_95ad18c19f", "text": "Print the token count for each chunk so I can size them.", "target_terms": ["token count"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_9a4f49c1fe", "text": "Can the status line show how much context is left?", "target_terms": ["status line"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_a1902dd6cd", "text": "The CLAUDE.md in this repo is getting too long, trim it down.", "target_terms": ["CLAUDE.md"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_a4a76a7f92", "text": "Drop the old cursorrules file once you migrate those conventions to the new format.", "target_terms": ["cursorrules"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": "fixed: Formulaic \"<Tool> reads <file> at the start\" clone and repeats \"house style\".", "term_origins": ["daniel"]} |
| {"id": "p_a4ddba3853", "text": "What reasoning effort is Codex set to in my config?", "target_terms": ["reasoning effort"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_a97425e01b", "text": "The diff is way bigger than it should be, did the formatter run on everything?", "target_terms": ["diff"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_bf0b103f9a", "text": "Run the untrusted script in the sandbox, not on the host.", "target_terms": ["sandbox"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_c42849956e", "text": "Check the token count before sending, because the long transcripts blow past the limit.", "target_terms": ["token count"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_c74d451c11", "text": "We're switching inference provider next month, so please test the new endpoint.", "target_terms": ["inference provider"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_c81568171e", "text": "Ask Codex for a code review of the pull request as a second opinion.", "target_terms": ["code review"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_cae8c84d98", "text": "Can you rewind the code but keep the conversation?", "target_terms": ["rewind"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_cbfdbc04b4", "text": "Expose reasoning effort as a flag in the script so I can tune it per job.", "target_terms": ["reasoning effort"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": "fixed: Textbook definition copying pack meaning.", "term_origins": ["claude"]} |
| {"id": "p_ccc36faced", "text": "I wrote a CLAUDE.md for the pipeline repo with the naming conventions.", "target_terms": ["CLAUDE.md"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_cf38710ab4", "text": "Keep the testing steps in CLAUDE.md, not in the chat, so they survive restarts.", "target_terms": ["CLAUDE.md"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_cfed9b1558", "text": "The token count for this file is roughly four times the size of the summary.", "target_terms": ["token count"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_d2917868ef", "text": "Update the tool schema so the search argument is required.", "target_terms": ["tool schema"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_d7dcd33655", "text": "Build the review bot on the Agent SDK so it gets the same tools as Claude Code.", "target_terms": ["Agent SDK"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_e3c5be4194", "text": "Validate the tool schema before starting the server so bad definitions fail fast.", "target_terms": ["tool schema"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": "fixed: Textbook definition copying pack meaning verbatim.", "term_origins": ["daniel"]} |
| {"id": "p_e402921a69", "text": "Hold back a validation set from every batch so we can see overfitting early.", "target_terms": ["validation set"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_e4e0b8665d", "text": "List every deferred tool from the gateway that mentions storage.", "target_terms": ["deferred tool"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_ec5862e66e", "text": "That's a deferred tool, so load its schema before you call it.", "target_terms": ["deferred tool"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_ed91087e1f", "text": "We rebuilt the support bot around retrieval-augmented generation so answers cite our own docs.", "target_terms": ["retrieval-augmented generation"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["daniel"]} |
| {"id": "p_ef4e955c11", "text": "Raise the reasoning effort for the migration plan, it has too many edge cases.", "target_terms": ["reasoning effort"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_fabcdf143a", "text": "I left the reasoning effort high by mistake, so the answer took ages.", "target_terms": ["reasoning effort"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
| {"id": "p_fd66b9dcac", "text": "Show me the diff before you commit.", "target_terms": ["diff"], "categories": ["tech", "ai_interactions"], "homophone_set": null, "split": "test", "derivation": "generated", "sentence_author": "claude (Claude Code subagent)", "source_dataset": null, "source_licence": null, "source_url": null, "source_authors": [], "source_date": null, "original_text": null, "review_note": null, "term_origins": ["claude"]} |
|
|
