Daohang Shi
Software Engineer at Facebook
United States
Summary
🤩
Rockstar🎓
Top SchoolDaohang Shi is a Staff GPU performance and systems engineer with 11 years of experience, currently driving low-level GPU tooling and performance work at Facebook. He brings deep expertise in CUDA and Triton and a strong track record optimizing ML frameworks, having contributed backend performance fixes and profiling enhancements to the widely used pytorch/pytorch project. With a PhD in Electrical Engineering from UW–Madison and a BS in Microelectronics from Fudan, he bridges rigorous academic training and production-grade systems engineering. Notably, he has improved Inductor compiler observability and unit-test robustness—work that helps teams diagnose and speed up model compilation and execution at scale.
11 years of coding experience
3 years of employment as a software developer
PhD, Electrical engineering, PhD, Electrical engineering at University of Wisconsin-MadisonBS, Microelectronics, BS, Microelectronics at Fudan UniversityHigh School Affiliated to Fudan University
Github Skills (18)
debugging10
pytorch10
debug10
performance-analytics10
performance-monitor10
python10
code-profiling10
performance-analysis10
gpu10
profiling10
performance-monitoring10
induction10
c-language9
machine-learning9
deep-learning9
Programming languages (2)
Github contributions (5)
pytorch/pytorch
Jan 2023 - Jan 2023
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:17 reviews, 1 commit, 21 PRs in 1 day
Contributions summary:Daohang primarily contributed to the PyTorch backend, focusing on improving performance and debugging. They fixed formatting issues in error messages and implemented logging for compilation times within the Inductor compiler. Furthermore, the user added regex matching to unit tests for the Inductor framework and addressed a key error related to pre-gradient passes. The user also made several updates to CProfile and time reports, contributing to the monitoring and profiling capabilities of the system.
pythongpu-accelerationdeep-learninggpunumpy
dshi7/pytorch
Jan 2023 - Jan 2025
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Contributions:76 pushes, 27 branches in 2 years
pythongpu-accelerationdeep-learninggpuacceleration
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.