About me

I am a PhD student at the University of Cambridge, supervised by Prof. Mateja Jamnik and Prof. Timothy Gowers, and a Student Researcher at Google DeepMind.

My research explores how large language models can learn to reason through reinforcement learning and make discoveries autonomously. I investigate these questions through applications to informal and formal mathematics, which I see as a rich testing ground for developing general methods for AI reasoning. My work spans LLM pre-training (OpenWebMath, Llemma), post-training for formal reasoning (Kimina-Prover), and agentic systems for autonomous theorem proving (Numina Lean Agent) and discovery.

I have long been passionate about mathematics and have been fortunate to work on AI for mathematics during research internships at OpenAI and Google DeepMind. More broadly, I am interested in the foundations and structure of mathematics: the connections between theories, the role of axioms, and the limits of proof and computation.

My CV, a description of my publications, and an overview of my research interests are available on this website.