About Me

I am an Assistant Professor in the Department of Computer Science at the University of Illinois Urbana-Champaign (UIUC). Prior to UIUC, I worked at Systems Research@Google and Meta AI, focusing on practical systems for GenAI.

Research Interests: Machine Learning & Systems, Cloud Computing, and Networked Systems

Our research brings together ML, systems, and computer networking to develop practical systems for efficient ML and data analytics, from the cloud to the planetary scale. Our systems have been adopted by big companies including Meta, LinkedIn, and Cisco. Our current topics of interests include:

  • Efficiency of Cloud ML Systems and Algorithms (e.g., LLMs, Image/VideoGen, DLRMs)
  • Systems for Collaborative Learning (e.g., federated learning, edge-cloud ML, hyperscale training)
  • (Systems for) ML for Systems

Prospective student: I am looking for highly self-motivated MS & PhD students and (remote) interns who are excited about ML and Systems research. Join us to work on exciting projects together!

News

  • [02/2025] Venn (On-device AI) accepted to MLSys'2025
  • [01/2025] (Fall) UIUC List of Teachers Ranked as Excellent (CS 598)
  • [11/2024] Honorable Mention for ACM SIGOPS Doctoral Dissertation Award!
  • [09/2024] LTE (LLM sparsity) accepted to NeurIPS'2024
  • [08/2024] Sequoia (distributed ML) accepted to SIGMOD'2025
  • [02/2024] FedTrans (personalized on-device ML) accepted to MLSys'2024
  • [10/2023] Wins David J. Kuck CSE Dissertation Prize!
  • [09/2023] Auxo (clustered federated learning) accepted to SoCC'2023
  • [05/2023] Selected as ML and Systems Rising Stars!
  • [03/2023] AdaEmbed (efficient large model pruning) accepted to OSDI'2023!
  • [03/2023] Fed-ensemble (personalized federated learning) accepted to IEEE ASE'2023
  • [03/2023] Win the Towner Prize for Distinguished Academic Achievement!
  • [01/2023] CCS (efficient ML training) accepted to ICLR'2023
  • [01/2023] Egeria (efficient ML training) accepted to EuroSys'2023

Recent Publications

(Full List)

Professional Service

  • Program Committee
    • 2026: ASPLOS, EuroSys
    • 2025: OSDI, NSDI, EuroSys, ICDCS
    • 2024: ATC, APSys, HotInfra
    • 2023: DistributedML, FL workshop@ICML
  • Reviewer for NeurIPS, ICML, CVPR

Teaching

Students

I am very fortunate to work with many wonderful students (incomplete list):

Prior Students at U. Michigan:
  • Xiangfeng Zhu (UM BSc → UW PhD student)
  • Yinwei Dai (UM MSc → Princeton PhD student)
  • Yile Gu (UM MSc → UW PhD student)
  • Yuxuan Zhu (UM BSc → UIUC PhD student)
  • Chengsong Zhang (UM BSc → UIUC MSCS student)
  • Xuye He (UM BSc → CMU Master's)