Li Li
Sinior Member Of Technical Staff at AMD
Austin, Texas, United States
Summary
🤩
Rockstar🎓
Top SchoolLi Li is a Senior Member of Technical Staff at AMD in Austin with nine years of experience spanning image processing, deep learning, control and power systems. A Ph.D. candidate in Signal and Image Processing at The George Washington University, he brings research rigor to production ML engineering, notably optimizing Facebook’s FBGEMM library for ROCm to improve embedding performance on AMD hardware. His background in control and power systems led to hands-on automation and C++ tooling for wind farm operations and large-scale PLC/server maintenance in hydropower, giving him an uncommon combination of hardware, firmware and ML software expertise. He has taught microcontroller, FPGA and electronics labs, demonstrating an ability to translate complex concepts into practical learning experiences. Comfortable moving between low-level performance tuning and high-level model development, he focuses on cross-platform efficiency and real-world deployment constraints.
9 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Signal and Image Processing, Doctor of Philosophy (Ph.D.) Signal and Image Processing at The George Washington UniversityBachelor's degree Automation (Control Engineering), Bachelor's degree Automation (Control Engineering) at Jilin University
Chinese, English
Github Skills (14)
pytorch10
machine-learning10
roc10
fba10
f10
performance-optimization10
emm10
cprogramming-language9
build-system9
c-language9
cmake9
hip9
cuda8
testing8
Programming languages (5)
Github contributions (5)
pytorch/FBGEMM
May 2022 - Sep 2022
FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/
Role in this project:
ML Engineer
ML EngineerContributions:66 reviews, 7 commits, 36 PRs in 3 months
Contributions summary:Li contributed significantly to enabling and optimizing the FBGEMM library for AMD devices, specifically ROCm. Their work included modifying build configurations for ROCm, fixing import issues, and enhancing the test suite to support ROCm. Furthermore, they made adjustments to the code, like changing hardcoded values, to improve the efficiency of the embeddingBag operation on ROCm architecture. These changes demonstrate a focus on performance and cross-platform compatibility within the machine learning domain.
matrix-multiplicationfacebookmultiplicationmatrixml-applications
ROCm/FBGEMM
May 2022 - Mar 2025
FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/
Contributions:8 reviews, 88 PRs, 224 pushes in 2 years 11 months
matrix-multiplicationlinear-algebrafacebookmultiplicationmatrix
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.