Presenting Claude Skills at a WorkOS company onsite
San Francisco · 2025AI-native DX. In practice.
I build the systems and teach the workflows that help technical teams turn AI from an experiment into how work gets done.
Work from the field.
Notes from building with AI daily, designing the systems around it, and teaching teams what survives contact with real work.
Browse all writing →Inside the Harness: Four Months of Being Wrong About pi
I tried pi in March, talked a coworker out of it in July, and rebuilt my entire agent stack on it in August. The thing I'd missed was that everything I'd been building around my harness could live inside it.
Read → claude-codeFleet: A Control Tower for Claude Code
A tmux dashboard for running many Claude Code agents at once — it shows which one needs you, sorted by urgency, so you stop polling and start responding.
Read → aiAIE Europe
A talk, a workshop, and a podcast at AIE Europe. Here's what I did.
Read → aiCase Statement: Building a Harness
I codified my job as a DX engineer into a system that dispatches agents, enforces conventions mechanically, and learns from its own failures. Here's how case works and why I think every developer should build something like it.
Read → aiWriting My First Evals
I had no background in evals. I built two very different evaluation systems for two AI-powered developer tools, and they taught me the same lesson: trust isn't a feeling, it's a measurement.
Read → aiIdeation: Because Planning Needs More Than a Mode
Claude Code's plan mode is a great starting point for thinking before coding. But for complex work, I needed more than a mode. I needed a system. Here's what I built.
Read →On stage and in the room.
Talks, workshops, and live demonstrations from a long-running public practice.
