I am a postdoctoral scholar in Computer Science at UC Berkeley, working with Prof. Ken Goldberg in the Berkeley AUTOLab, part of Berkeley AI Research (BAIR). I received my Ph.D. from UC Berkeley in 2025, co-advised by Prof. John Kubiatowicz and Prof. Ken Goldberg, and my B.S. summa cum laude from UCLA in 2020, where I worked with Prof. Songwu Lu on wireless networking.

I build the systems that let robots learn and act at scale, and most recently the agentic tooling that lets LLM coding agents program, simulate, and improve robots. I co-lead Graph-as-Policy (GaP) (CoRL 2026 Spotlight, top 3.4%), a multi-agent self-learning harness in which LLM agents write, verify, and optimize executable behavior graphs for variational automation tasks, and RoboSimStudio, an interactive toolkit that lets humans and coding agents rapidly build Real2Sim simulations of robot manipulation tasks — both in collaboration with NVIDIA.

My earlier work evaluates and trains vision-language models with robot data — RoboVista (RSS 2026) and Robo2VLM (NeurIPS 2025 Spotlight) — and builds cloud-robotics infrastructure with FogROS2, which ships as part of the official ROS 2 release. Previously, I was one of six core developers of SkyPilot, and spent summers at Bosch Research and VMware Research.