What I'm up to

My main focus right now is AI-powered code analysis tooling. I shipped claude-audit— a CLI that analyzes your entire codebase for security vulnerabilities using a multi-agent architecture with specialized domain agents. It's on npm, has 6 stars and growing.

Working on llm-ocr — converting PDFs to clean markdown using vision LLMs instead of traditional OCR. The results are surprisingly good for documents with mixed layouts.

Also building this very site — designed to be machine-readable by AI models while still being a good experience for humans. Check /llms.txt.

How LLMs can be better tools for developers rather than replacements. The gap between AI hype and practical utility is still enormous. Most AI features in products today feel bolted on. I want to figure out the patterns that make AI genuinely useful.

Interested in multi-agent systems that actually work in production — not just impressive demos but reliable, testable architectures that handle edge cases gracefully.

Developer experience as a design discipline. Good DX isn't just documentation — it's the API surface, error messages, sensible defaults, and respecting the developer's time.

Advanced agent frameworks and orchestration patterns. How to build agents that are reliable enough for production use, not just hackathon demos.

Systems design at scale — understanding how the tools I build would behave under real load and with real constraints.

The craft of writing clearly about technical ideas. Explaining complex systems in simple language is underrated and incredibly hard.