Zhuoran Yin
Senior Member Of Technical Staff at AMD
Chapel Hill, North Carolina, United States
Summary
🤩
Rockstar🎓
Top SchoolZhuoran Yin is a system-level software engineer with 10 years of experience specializing in deep learning compilers and GPU backend performance. As a Senior Member of Technical Staff at AMD, he led MLIR/MIOpen integration and implemented/optimized implicit GEMM algorithms to accelerate convolution on ROCm platforms. His open-source contributions to projects like XLA, IREE, and MIOpen show deep expertise in GPU code generation, MFMA tuning, and layout-flexible tensor APIs—often focused on making AMD GPUs first-class citizens in major ML compilers. Based in Chapel Hill, he combines production-focused engineering with research roots from a MASc at University of Waterloo, and is known for pragmatic compiler heuristics that boost real-world kernel performance for narrow, challenging problem sizes.
10 years of coding experience
9 years of employment as a software developer
Bachelor of Engineering (B.Eng.), Bachelor of Engineering (B.Eng.) at Hunan UniversityMaster of Applied Science (MASc) - Thesis, Master of Applied Science (MASc) - Thesis at University of Waterloo
English, Chinese
Github Skills (29)
performance-analytics10
c-language10
performance-monitor10
operation10
llvm10
tensorrt10
gpu-programming10
machine-learning10
mlr10
performance-measurement10
roc10
compiler-compiler10
performance-analysis10
tensorflow10
gpu10
Programming languages (8)
Github contributions (5)
ROCm/MIOpen
Mar 2020 - Jan 2023
AMD's Machine Intelligence Library
Role in this project:
Back-end Developer
Back-end DeveloperContributions:382 reviews, 57 commits, 48 PRs in 2 years 10 months
Contributions summary:Zhuoran contributed significantly to the core functionality of the MIOPEN library. They focused on optimizing and refactoring existing kernels, as evidenced by the changes in `gridwise_gemm` and various kernel algorithm files. They also implemented flexible layout support within the tensor API, including modifying descriptors and updating problem descriptions. Furthermore, the user added support for MLIR-cpp backend with a focus on implementing and tuning solvers, demonstrating a strong understanding of performance optimization.
amdcluster-computingmachine-learningblasmachine-intelligence
openxla/xla
Jun 2019 - Aug 2020
A machine learning compiler for GPUs, CPUs, and ML accelerators
Role in this project:Back-end Developer
Back-end DeveloperContributions:42 commits in 1 year 1 month
Contributions summary:Zhuoran's commits primarily focus on adding and modifying code related to the ROCm (AMD GPU) platform within the XLA (XLA:GPU) compiler. This includes implementing compatibility features for ROCm GPUs, specifically addressing the `gpu_executable` and related components. The user also contributed to fixing CUDA compilation issues, which suggests a strong understanding of GPU code generation and compilation pipelines within the project. The user updated dependent symbols and comments as well.
compilercommunity-drivenmachine-learningmodular
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.