ML Engineer (ex-FAANG). Building the Trust & Reliability layer for AI.

My current project is Steer, an open-source library for deterministic AI agent reliability.

GitHub: https://github.com/imtt-dev/steer

I write about the 'Confident Idiot' problem and the shift from passive monitoring to active teaching.

Blog: https://steer-labs.com