Pieter Abbeel is a professor of electrical engineering and computer sciences at UC Berkeley, where he directs the Robot Learning Lab and co-directs the Berkeley Artificial Intelligence Research (BAIR) Lab. Born in Belgium, he earned his Ph.D. in computer science at Stanford in 2008 and has spent his career making robots learn rather than be hand-programmed.
Abbeel’s research spans apprenticeship and imitation learning (robots learning from human demonstrations), deep reinforcement learning for robotic control, and meta-learning, the idea of training systems to learn new tasks quickly. He advised a generation of influential researchers, including John Schulman on the TRPO and PPO algorithms, and co-authored landmark work such as the denoising diffusion probabilistic models paper that helped launch modern image generation. He received the 2021 ACM Prize in Computing for contributions to robot learning.
Beyond academia he has founded several companies, including Gradescope (AI-assisted grading), Berkeley Open Arms (low-cost robot arms), and Covariant, a startup applying AI to warehouse and factory robotics. In 2024 Amazon licensed Covariant’s robotics foundation models and hired its founders, and Abbeel later took a leadership role in Amazon’s AI organization while continuing his Berkeley work.