|
Zezhou Wang
I am a Ph.D. student in Computer Science at the
Paul G. Allen School, University of Washington,
advised by Prof. Arvind Krishnamurthy.
My research interests lie in computer systems and machine learning systems,
with a current focus on enabling flexible compute scheduling in distributed LLM systems.
During my internship at ByteDance Seed,
I co-developed veScale-FSDP, which supports
structure-aware training methods, such as block-wise quantization and matrix-based optimizers,
through a flexible RaggedShard DTensor abstraction.
Email /
Google Scholar /
CV /
GitHub
|