Hi, I’m Chenxi Yang, a Research Scientist at Meta MSL. I focus on serving systems for foundation models at scale, with a recent emphasis on multimodal systems. My interests lie in co-design across models, infrastructure, and applications.

I received my Ph.D. in Computer Science from the University of Texas at Austin, advised by Swarat Chaudhuri, and my B.S. from Fudan University, where I worked with Yang Chen and Chenren Xu. During my early graduate studies, I also worked with Lili Qiu on video streaming systems for DNNs.

During my Ph.D., I developed theory and tools (DSE, CAROL, Canopy) that provide performance and robustness guarantees for ML systems. I extended this line of work to large-scale production environments through two internships at Google, focusing on TPU optimization (SRG) and ML-based storage optimization, working with Yawen Wang, Elliot Li, Mustafa Uysal, and Martin Maas.

Outside of work, I enjoy reading, playing tennis, hiking, and travelling.