I am a Member of Technical Staff at Reflection AI working on reasoning and agentic RL for post-training. Previously, I was a Senior Staff Research Scientist at Google DeepMind, where I helped to develop RL-based reasoning methods for Gemini 2.0 Flash Thinking and Gemini 2.5. My research has also focused on model-based reinforcement learning, structured relational architectures, and cognitive science.

I completed my PhD in Psychology at UC Berkeley in Tom Griffiths’ Computational Cognitive Science Lab. Before that, I researched intuitive physics in Josh Tenenbaum’s Computational Cognitive Science Group at MIT.

During my PhD, I was also heavily involved in Project Jupyter, and was honored to receive the 2017 ACM Software System Award along with the other members of the Jupyter Steering Council. I also created a tool called nbgrader for developing homework assignments in the Jupyter Notebook (though I am no longer a maintainer).

I like flowers 🌻, board & video games 🎮, and spending time outside ⛰️.

Resume Academic CV