Spaces:
Running
Running
fix: add README.md with frontmatter for static Space
Browse files🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
README.md
CHANGED
|
@@ -7,88 +7,4 @@ sdk: "static"
|
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
**AI Research & Engineering · Multi-Agent System Coordination**
|
| 13 |
-
|
| 14 |
-
I research how AI agents coordinate at scale and build production tools to make it practical.
|
| 15 |
-
|
| 16 |
-
---
|
| 17 |
-
|
| 18 |
-
## Research
|
| 19 |
-
|
| 20 |
-
**Harmonic Coordination Theory (HCT)** proposes a musical ontology for multi-agent coordination. Current frameworks give you orchestration tools but lack a shared language for timing, quality, and intent.
|
| 21 |
-
|
| 22 |
-
I fix this using cues, fermatas, tempo, and dissonance as engineering primitives.
|
| 23 |
-
|
| 24 |
-
### Key Publications
|
| 25 |
-
- [Harmonic Coordination Theory](https://www.stefanwiest.de/research) — CS.AI / CS.MA papers
|
| 26 |
-
- [stefanwiest/hct-spec](https://huggingface.co/stefanwiest/hct-spec) — Canonical signal specification
|
| 27 |
-
- [stefanwiest/hct-core](https://github.com/stefanwiest/hct-core) — Reference implementation
|
| 28 |
-
|
| 29 |
-
### Protocol Extensions
|
| 30 |
-
- [stefanwiest/hct-mcp-signals](https://github.com/stefanwiest/hct-mcp-signals) — Coordination layer for Anthropic MCP
|
| 31 |
-
- [stefanwiest/hct-a2a](https://github.com/stefanwiest/hct-a2a) — Coordination semantics for Google A2A
|
| 32 |
-
|
| 33 |
-
---
|
| 34 |
-
|
| 35 |
-
## Tools
|
| 36 |
-
|
| 37 |
-
Production tools that solve real problems—usually ones I've encountered myself.
|
| 38 |
-
|
| 39 |
-
| Tool | Purpose | Link |
|
| 40 |
-
|------|---------|------|
|
| 41 |
-
| **AntiSlop** | Multi-language linter for AI-generated code slop | [GitHub](https://github.com/skew202/antislop) |
|
| 42 |
-
| **EnvCheck** | Environment file linting for DevSecOps | [GitHub](https://github.com/envcheck/envcheck) |
|
| 43 |
-
| **NerfStatus** | Scientific LLM degradation detection | [Web](https://nerfstatus.com) |
|
| 44 |
-
| **SpeakOps** | Real-time AI speech coaching | [GitHub](https://github.com/speakops/desktop) |
|
| 45 |
-
|
| 46 |
-
---
|
| 47 |
-
|
| 48 |
-
## Writing
|
| 49 |
-
|
| 50 |
-
[**SeekingSota**](https://seekingsota.com) — Weekly essays on AI × Human identity crisis
|
| 51 |
-
|
| 52 |
-
I explore what happens when engineers stop programming and start conducting AI agents. Real agent workflows, technical deep dives, philosophy of production, and honest stories about the identity crisis that comes with this transformation.
|
| 53 |
-
|
| 54 |
-
---
|
| 55 |
-
|
| 56 |
-
## HuggingFace
|
| 57 |
-
|
| 58 |
-
**Models & Datasets** (`stefanwiewst/`):
|
| 59 |
-
- [stefanwiest/hct-spec](https://huggingface.co/stefanwiest/hct-spec) — HCT specification & signal reference
|
| 60 |
-
|
| 61 |
-
**Spaces** (`skew202/`):
|
| 62 |
-
- [skew202/nerfstatus-hf-monitor](https://huggingface.co/spaces/skew202/nerfstatus-hf-monitor) — HF inference quality monitor
|
| 63 |
-
|
| 64 |
-
---
|
| 65 |
-
|
| 66 |
-
## Collections
|
| 67 |
-
|
| 68 |
-
Curated research collections grounding multi-agent systems in academic literature:
|
| 69 |
-
|
| 70 |
-
- **[Multi-Agent Coordination & Signaling](https://huggingface.co/collections/stefanwiest/multi-agent-coordination-and-signaling-6957724f871e3a888d717ce4)** — A2A, MCP, CAMEL, IoA, HCT protocols
|
| 71 |
-
|
| 72 |
-
- **[LLM Reasoning & Planning Techniques](https://huggingface.co/collections/stefanwiest/llm-reasoning-and-planning-techniques-69577251137782117c4573ec)** — CoT, ToT, GoT, ReAct, DSPy
|
| 73 |
-
|
| 74 |
-
- **[LLM Quality & Degradation Monitoring](https://huggingface.co/collections/stefanwiest/llm-quality-and-degradation-monitoring-69577255871e3a888d717d1d)** — Hallucination detection, model collapse, probe-based testing
|
| 75 |
-
|
| 76 |
-
- **[Memory, Context & RAG](https://huggingface.co/collections/stefanwiest/memory-context-and-rag-69577253871e3a888d717d09)** — RAG architectures, MemGPT, semantic chunking
|
| 77 |
-
|
| 78 |
-
- **[Tool Use & Agent Execution](https://huggingface.co/collections/stefanwiest/tool-use-and-agent-execution-69577257c4fed2c5f572e482)** — Toolformer, MCP, agentic loops
|
| 79 |
-
|
| 80 |
-
- **[Agent Safety & Alignment](https://huggingface.co/collections/stefanwiest/agent-safety-and-alignment-69577259df172c4fd94b5a34)** — Jailbreak detection, guardrails, constitutional AI
|
| 81 |
-
|
| 82 |
-
---
|
| 83 |
-
|
| 84 |
-
## Connect
|
| 85 |
-
|
| 86 |
-
- 🌐 [Website](https://stefanwiest.de)
|
| 87 |
-
- 💼 [LinkedIn](https://linkedin.com/in/stefanwiest)
|
| 88 |
-
- 🐦 [X/Twitter](https://x.com/seekingsota)
|
| 89 |
-
- 🤗 [Hugging Face](https://huggingface.co/stefanwiest)
|
| 90 |
-
- 📦 [GitHub](https://github.com/skew202)
|
| 91 |
-
|
| 92 |
-
---
|
| 93 |
-
|
| 94 |
-
*15+ years building production software. Now researching multi-agent coordination while building tools and writing honest stories about the transformation.*
|
|
|
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
+
Stefan Wiest - AI Research & Engineering
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|