I am an Assistant Professor in the Machine Learning Department and Computer Science Department at Carnegie Mellon University . I am also a Distinguished Engineer at NVIDIA. I was the Chief Technologist of OctoAI(acquired by NVIDIA). I recently received my Ph.D. from Paul G. Allen School of Computer Science & Engineering, University of Washington.

At UW, I was advised by Carlos Guestrin. I also worked closely with Luis Ceze and Arvind Krishnamurthy. I was a member of SAMPL Lab and MODE Lab. I recieved my Master and Bachelor's degree from Prof. Yong Yu's ACM Honors Class at Shanghai Jiao Tong University (SJTU).

Research

I am interested in the intersection of machine learning and systems. I like to bring a whole stack view from models, systems down to the low-level hardware backend to solve real world AI and systems problems. We are working on exciting problems across the spectrum together at CMU Catalyst Group.

I am a believer of open source and open science. My group involves publishing algorithms in openly accessible mediums and building open-source machine learning systems that are widely adopted. Here are the ML systems that I created:

• MLC-LLM Universal LLM Deployment Engine with ML Compilation
  
• Apache TVM, an Automated End-to-End Optimizing Compiler for Deep Learning.
  
• XGBoost, a scalable tree boosting system.
  
• Apache MXNet(co-creator)
  

Activity and Services

• Board President, MLSys Conference.
• PC Chair, MLSys 2023.
• Artifact Evaluation Chair, MLSys 2022
• PC member of MLSys.
• Area Chair, ICML, NeurIPS.

Information

• Prospective students FAQ

Experience

• CMU, present
  Assistant Professor

• NVIDIA, present
  Distinguished Engineer

• OctoAI, 2019 - 2024
  Chief Technologist.

• University of Washington, 2013 - 2019
  Ph.D. student with Carlos Guestrin.