Yi Zhang
Software Engineer II at Microsoft
Suzhou City, Jiangsu, China
Summary
🤩
Rockstar🎓
Top SchoolYi Zhang is a Software Engineer II at Microsoft with 10 years of experience building resilient backend and CI/CD systems, especially for ML and GPU-focused projects. He has deep expertise automating builds and testing across Windows and Linux, contributing to highly visible open-source repositories such as PyTorch and ONNX Runtime where he fixed dependency, packaging, and CUDA-related build issues. Comfortable across full-stack and DevOps roles, he has shipped production tooling for distributed training (DDP) and improved cross-platform developer workflows. His background spans enterprise AI at Zelros, telecom automation, and frontend work at SAP, reflecting a pragmatic blend of research-grade ML infrastructure and product-focused engineering. Notably, he combines hands-on Docker/Dockerfile and Windows build problem-solving with an engineer’s attention to testing and reproducibility.
10 years of coding experience
4 years of employment as a software developer
Engineer's degree, Information and System, Engineer's degree, Information and System at Université de Technologie de TroyesBachelor's degree, Information Engineering, Bachelor's degree, Information Engineering at Shanghai University
English, French, Chinese, Chinese
Github Skills (43)
pytorch10
docker10
back-end-development10
python10
windows10
ddp10
data-parallel10
machine-learning10
bash10
build-system10
batchfile10
dockers10
cicd10
compatibility10
data-loading10
Programming languages (9)
Github contributions (5)
microsoft/onnxruntime
Dec 2019 - Jan 2023
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:
Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:469 reviews, 358 commits, 620 PRs in 3 years 1 month
Contributions summary:Yi primarily contributed to the build and CI/CD aspects of the onnxruntime repository. Their work included fixing build issues related to the ngraph wheel and updating build scripts, specifically the Dockerfile and related scripts for the Linux build. The user also made changes to the testing infrastructure, adding new builds for different architectures, and addressed packaging pipeline issues. The user demonstrated proficiency in build automation and resolving dependencies.
runtimetrainingtensorflowai-frameworkaccelerator
pytorch/builder
Nov 2020 - Jan 2022
Continuous builder and binary build scripts for pytorch
Role in this project:DevOps Engineer & Automation Engineer
DevOps Engineer & Automation EngineerContributions:15 reviews, 11 commits, 89 PRs in 1 year 2 months
Contributions summary:Yi primarily contributed to the build and continuous integration/continuous delivery (CI/CD) infrastructure of the pytorch/builder repository. Their work involved modifying build scripts, including those for Windows, to address issues like CUDA installation and Windows update handling. They also focused on automating the download and integration of dependencies such as Miniconda, along with improving the testing framework by adding checks for CUDA and CNN functionality.
pytorchbuild-scriptscontinuousbuilder
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.