Now

Currently at Meta, working on AI systems for internal operations teams. Most of my time goes into LLM-based tooling and OR models that have to actually work in production, not just demo well. The cost of being wrong there is real, and I like that.

On the side, I'm deep in sarvis, an iOS project trying to make your phone smarter about what you're thinking, not just what you're doing.

I'm also spending more time thinking about how AI agents should be structured when the task is open-ended. No clean answers yet.