whoami

Building has always been the way I make sense of things.

This is where I leave notes, experiments, findings, and occasional opinions from the systems I'm pulling apart or putting together. Lately, many of those systems happen to involve AI.

I work on the boundary between model behavior and system behavior: what the model gets wrong, what the harness makes worse, and what can actually be made reliable.

Mostly writing about LLM harnesses, local deployments, open source, ways of working, and the occasional strange connection between otherwise unrelated fields.

Some people write, paint, or compose. Code turned out to be my medium, and I'm happiest when it becomes systems with moving parts. That's probably the closest thing I have to a canvas.

If any of this touches your own work, send me a note.