Changming Sun
San Francisco, California, United States
Summary
🤩
RockstarChangming Sun is a Principal Software Engineer in San Carlos, CA with 12 years of experience and a leading role on ONNX Runtime since 2018, now driving build systems, core framework, security & compliance, and packaging and release engineering at Microsoft. He combines systems- and infrastructure-level expertise with ML inference know-how, focusing on cross-platform, high-performance runtime behavior. His open-source work spans deep backend improvements in ONNX and ONNX Runtime—fixing LSTM/concat ops and shape inference, hardening tests, eliminating memory leaks, and even converting Eigen into a submodule to simplify builds. He also pays attention to developer-facing details, contributing CI-integrated inference examples and adapting image-processing examples to Windows Imaging Component, demonstrating end-to-end stewardship from low-level code to consumable samples. With a B.S. in Information and Computing Science, he brings a pragmatic blend of engineering rigor and product-minded execution.
12 years of coding experience
Stackoverflow
917reputation
618kreached
27answers
5questions
Github Skills (27)
optimizations10
c-language10
onnx-runtime10
python10
image-processing10
machine-learning10
cmake10
onnx10
deep-learning10
optmization10
performance-optimization10
optimisation10
performance-tuning10
c-programming-language10
optimization10
Programming languages (23)
Github contributions (5)
microsoft/onnxruntime
Nov 2018 - Dec 2022
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:
Backend Engineer
Backend EngineerContributions:5 releases, 4537 reviews, 1142 commits in 4 years 1 month
Contributions summary:Changming's commits primarily focused on enhancements and bug fixes related to the performance test runner within the ONNX Runtime project. They added functionality to build the test runner conditionally, fixed memory leaks, and addressed a bug within the runner's core logic. The user also converted the Eigen library to a submodule and updated to the latest version.
runtimetrainingtensorflowai-frameworkaccelerator
microsoft/onnxruntime-inference-examples
Jul 2021 - Oct 2022
Examples for using ONNX Runtime for machine learning inferencing.
Role in this project:ML Engineer
ML EngineerContributions:94 reviews, 36 commits, 106 PRs in 1 year 3 months
Contributions summary:Changming contributed to the example code for machine learning inferencing with ONNX Runtime. Their commits included adding and updating example code, specifically related to image processing with JPEG files and model updates. The user also focused on integrating the example into a CI build pipeline and updating the ONNX Runtime version used. Further changes involved adapting the fns_candy_style_transfer example to use Windows Imaging Component (WIC).
deep-learningruntimeinferencemachine-learningonnx
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.