I build infrastructure that makes AI coding agents work better. Caveman optimizes agent output, Cavekit drives specification-driven development, and The Prompt Library gives agents structured access to production prompts — all open-source tools that share one thesis: AI agents need better tooling, not just better models.

I'm also a founding engineer at Stacklink, an enterprise RAG platform, and building Revu, a native macOS study app that applies spaced repetition and multi-agent AI to education. The common thread is systems that compound — retrieval that improves with use, review schedules that adapt to recall, agents that learn from specifications.

I study Data Science & AI at Leiden University, where I sit on the Education Committee representing students in curriculum development. I write about AI systems and indie development on my personal publication, Polder.