Kang-Fu Mei
梅康夫

Research Scientist and Tech Lead at Google DeepMind, working on reinforcement learning post-training for multimodal generative and world models.

My work spans reward modeling, data, evaluation, and offline and online RL. I’m interested in systems that can simulate our dynamic world for creative control, more robust autonomous systems, and new entertainment experiences.

I received my Ph.D. in Electrical and Computer Engineering from Johns Hopkins University, where I worked with Prof. Vishal M. Patel. Before that, I completed my M.S. at The Chinese University of Hong Kong, Shenzhen, advised by Prof. Rui Huang.