I'm a senior research engineer at Google DeepMind, working on reinforcement learning and large language models. I created the test-time RL (TTRL) inference method behind AlphaProof — the first AI system to reach (silver) medal-level performance at the International Mathematical Olympiad (Nature 2025). I'm passionate about pushing AI toward superhuman capabilities through novel algorithms, mathematical insights, and scalable engineering.

Beyond AlphaProof, I'm a co-first author of Formal Conjectures, an open benchmark for verified mathematical discovery, and contributed to AlphaProof Nexus, an AI-driven formal proof search that resolved nine open Erdős problems, and to DeepMind's IMO 2025 gold-medal effort. I've also worked on the Gemini 3 family and on SynthID-Image, image watermarking at internet scale.

I hold a Master's and Bachelor's degree in Computer Science from ETH Zürich. During my Master's, I conducted AI research at ETH Zürich and completed internships as a software developer at Amazon and a quantitative researcher at Jump Trading. I'm also a bronze medallist at the International Mathematical Olympiad (IMO 2016).