I am a theoretical physicist at Lawrence Berkeley National Lab, working at the intersection of deep learning, statistical mechanics, and mathematical physics. My research focuses on two areas: developing analytical tools for mechanistic interpretability—understanding how neural networks internally represent and process information—and building physically interpretable generative models for scientific applications, particularly for amorphous materials and protein and RNA folding

I completed my PhD in physics at UC Berkeley in 2025, where I developed new approaches to dark matter detection from a condensed matter perspective. My previous research has spanned condensed matter and materials physics, nonlinear dynamics and PDEs, and quantum algorithms. See below for an overview of my research and publications.

On a personal level, I enjoy building mechanical keyboards and reading. I live with my wife, dog, and two cats in the SF Bay Area.