About

AI/ML engineer focused on building systems that people can actually trust. My work spans healthcare AI (Synthure), sports analytics (Replays AI), and AI safety research (CORDA Fellow at NYU), and the common thread across all of it is using engineering and cloud infrastructure to solve problems that have real consequences. I care about the gap between a model that looks good on a benchmark and one that behaves correctly in the real world, and I spend a lot of time in that gap. My research interests sit at the intersection of AI safety, Bayesian methods, and production ML, with a focus on alignment drift, adversarial robustness, and catching failure modes before they reach users.

Outside of building things, I write about ML from first principles on my blog, covering optimization geometry, generalization theory, and probabilistic modeling with real benchmarks. I am also a competitive chess player and a devoted fan of the Golden State Warriors and San Francisco 49ers. When I am not building things I am grinding tactics on Lichess or watching game film.