About Me

I’m an AI researcher and technical leader interested in building intelligent systems that work reliably in the real world—and in figuring out how we know when they do.

My work spans AI research, cognitive science, and engineering, with a particular interest in reasoning, agent behavior, and evaluation. I like problems that start out underspecified: turning a research question or emerging capability into something concrete enough to build, measure, test, and improve. Over the course of my career, that has meant everything from studying how AI systems reason about people’s beliefs and goals to designing evaluation frameworks, building AI prototypes and research infrastructure, and leading interdisciplinary technical teams.

I’m especially drawn to work at the boundary between research and application: understanding what new AI capabilities make possible, identifying where they break down, and turning that understanding into better systems, tools, and products. Lately, I’ve been particularly interested in robust evaluation, AI reliability and safety, and the ways representation and context shape model behavior.

Looking for my old site? It’s archived here.