About Me

I’m a research engineer thinking about AI systems — how they reason, how they fail, and how we might come to understand them better. My background is in large-scale LLM infrastructure and reinforcement learning, and I’m increasingly focused on alignment research.

I’m inspired by work like AlphaEvolve — systems where AI doesn’t just answer questions but actively discovers, iterates, and extends what we know. I want to build AI systems in that spirit: ones that compound, improve themselves, and meaningfully push the frontier rather than just sit at it.

I’m drawn to questions that sit between disciplines: what pre-linguistic cognition might teach us about machine reasoning, how frameworks from philosophy of mind apply to interpretability, what prediction and Bayesian reasoning look like as habits of thought rather than just techniques.

I write occasionally about probability, risk, and strategy. I’m also interested in quantitative trading, the architecture of meaningful spaces, and the long arc of how technical work connects to human experience.

I collect watches. What I really collect, though, are stories — the history behind an object, the choices in its making, the precision someone insisted on when they didn’t have to. I find that kind of attention compelling wherever it shows up: in a movement, a building, a piece of writing, a well-played game.