Andi Peng

Andi Peng leads national security safety evaluations at Anthropic and is a PhD student at MIT. Her research has primarily focused on learning good representations and preference models for RL agents and robots.