Xieyuan (Francis) ZHANG

I am a researcher at Wayve Labs in London, where I work on (offline) reinforcement learning & generative policiesFor a glimpse into our work on RL, see this talk given by our CEO, Alex, at the CVPR 2026 Workshop on Autonomous Driving. for autonomous driving.

I have long been interested in control from pixelsA collection of great talks that have shaped my research perspective.. While my primary focus is end-to-end autonomous driving, closing the loop between observation and action is a challenge shared across many domains. I draw inspiration from progress in other fields, from contact-rich robotic manipulation ($\pi^\star_{0.6}$) to computer-use agents (Yutori Navigator).

Before joining Wayve, I earned an M.S. in Mechanical Engineering at Stanford University, specializing in automatic control. There, I developed a strong interest in optimal control and reinforcement learning, spanning both their theoretical foundations and applications in vehicle control and finance.