Yuan M
Engineering Manager at Intel Corporation
Shanghai, United States
Summary
🤩
Rockstar🎓
Top SchoolYuan M is an Engineering Manager at Intel with over 15 years of experience designing and optimizing system-level software for processor and GPU platforms. He progressed from hands-on senior engineering roles at Qualcomm to technical lead and now manages a team delivering core software products, recruiting and mentoring engineers while driving cross-functional execution. Yuan combines deep C++ backend and performance engineering expertise with architecture-level decision making, notably contributing to the Intel GPU backend of the widely used OpenVINO toolkit to resolve performance bottlenecks and hard-to-reproduce segfaults. He focuses on practical optimizations—such as MatMul dequantization and group normalization improvements—that directly improve AI inference performance in production. Based in Shanghai and educated at UC San Diego and UCLA, he blends academic grounding with enterprise-scale delivery. Colleagues rely on him to translate complex system problems into reliable, high-performing implementations and scalable team practices.
5 years of coding experience
15 years of employment as a software developer
University of California, San DiegoUniversity of California, Los Angeles
Github Skills (12)
gpu-programming10
c-language10
cprogramming-language10
gpu10
performance-optimization10
intel10
openvino10
ai9
deeplearning-ai9
inference9
deep-learning9
computer-vision9
Programming languages (4)
Github contributions (5)
openvinotoolkit/openvino
Jun 2024 - Feb 2025
OpenVINO™ is an open source toolkit for optimizing and deploying AI inference
Role in this project:
Backend Engineer & Performance Engineer
Backend Engineer & Performance EngineerContributions:18 reviews, 11 PRs, 32 comments in 7 months
Contributions summary:Yuan primarily contributed to optimizing the Intel GPU backend of the OpenVINO toolkit. Their work focused on resolving performance bottlenecks and fixing bugs related to the GPU backend, specifically addressing issues in MatMul dequantization, OneHot, and MVN operations. The contributions involved modifying C++ code within the Intel GPU plugin, including changes to primitive implementations and graph optimization passes, and involved addressing issues that resulted in segmentation faults. Furthermore, the user was involved in improving the performance of group normalization for a specific model.
inference-enginepytorchmodel-optimizerdeep-learninggpu
yuanxion/openvino
Jul 2024 - Mar 2025
OpenVINO™ is an open-source toolkit for optimizing and deploying AI inference
Contributions:1 PR, 98 pushes, 14 branches in 8 months
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.