Yuqing Xia
Senior Researcher at Microsoft
Haidian District, Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolYuqing Xia is a Senior Researcher at Microsoft in Beijing with seven years of software and ML engineering experience bridging research and production. Trained as a PhD in cell biology from Peking University, she brings a quantitative, experimental mindset to systems-level problems in deep learning compilers and runtime optimization. At Microsoft she progressed from software engineer to senior researcher, contributing notable backend and CUDA/ROCm support to the widely used nnfusion project and integrating Antares kernel emission for high-performance kernel generation. Her work touches low-level memory management, code generation, and cross-platform execution engines, making her equally comfortable with algorithmic research and pragmatic engineering. Based in Haidian District, she combines domain rigor from life sciences with hands-on compiler engineering to accelerate ML model deployment at scale.
7 years of coding experience
4 years of employment as a software developer
Bachelor of Science, Biotechnology, Bachelor of Science, Biotechnology at Sichuan UniversityDoctor of Philosophy (PhD), Cell Biology, Doctor of Philosophy (PhD), Cell Biology at Peking University
English
Github Skills (7)
cuda10
c-language10
cprogramming-language10
roc9
code-generation9
compiler-optimization9
machine-learning8
Programming languages (3)
Github contributions (5)
microsoft/nnfusion
Sep 2020 - Jan 2023
A flexible and efficient deep neural network (DNN) compiler that generates high-performance executable from a DNN model description.
Role in this project:
Backend Developer & ML Engineer
Backend Developer & ML EngineerContributions:64 reviews, 95 commits, 103 PRs in 2 years 4 months
Contributions summary:Yuqing's primary contribution revolves around the development and optimization of the deep neural network compiler. They focused on enhancing the compiler's engine by implementing CUDA and ROCm support. Key tasks included adding CPU engine, removing interpreter and backend manager, code formatting, and addressing architectural aspects related to memory management and code generation. They integrated Antares kernel emitter within the CUDA backend, thereby, enabling the compilation of optimized kernels for a high-performance runtime environment.
compilerneural-network
xiayuqing0622/tilelang
Mar 2025 - Oct 2025
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
Contributions:50 pushes, 4 branches in 7 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.