Published multiple papers such as AMP, DistFlashAttn on distributed training systems, and MPCFormer, Marill on distributed serving systems. Received awards including Amazon AI Fellowship (2025) and Nvidia Graduate Fellowship Finalist (2025-2026).
Research Experience
Affiliated with Sky Computing Lab and Berkeley AI Group (BAIR), co-leading NovaSky and an early member of lmsys. Interned at Nvidia Research and Google.
Education
Currently a third-year CS PhD student at EECS, UC Berkeley, advised by Prof. Ion Stoica and Prof. Joseph Gonzalez; also works closely with Prof. Song Han (MIT). Obtained a master's degree in Machine Learning from CMU, advised by Prof. Eric Xing and Prof. Hao Zhang; and a bachelor's degree with double majors in Computer Science and Math from UC San Diego, advised by Prof. Zhuowen Tu.
Background
Research interests include the efficient development of artificial intelligence and artificial worlds, particularly at the intersection of visual and textual generative models with distributed systems.
Miscellany
Personal interests and other information are not explicitly mentioned in the provided HTML content.