My work spans full-stack engineering, AI evaluation, and tools that help domain experts investigate data and test ideas. I’m especially interested in how people and AI can do better scientific work together.
How we built Insight Miner at Airbnb: an agent harness that helps domain experts investigate unstructured text and evaluate AI through shared, reproducible scientific methods.
Read the articleDynamic Stability · Essay
A practical guide to making AI a better collaborator in engineering and science: giving it the tools and context it needs, learning from its failures, and rethinking where your own effort belongs.
Read the essay
Foiling, racing, and life on the water.
Independent product
Foil Run started with algorithms I wrote to compare foil gear. It’s grown into my dream app for understanding our riding, bringing together measurements and riders’ own observations to study gear and performance. It’s also my benchmark project for new AI models and tools: I put each release to work on something I use and care about.
Case study
A specialist company’s product knowledge was scattered across emails, chats, and product documentation. I built a support tool that brings that knowledge together and drafts replies for expert review, preserving their judgment and voice.
How coordinating AI coding agents into a small fleet turned nights-and-weekends tinkering into shipped products, Foil Run among them.
Using coding agents to organize journal entries, review decisions, and revisit recurring patterns.
Journaling project on GitHub