Lei Cao
Senior Software Engineer at Microsoft
Redmond, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolLei Cao is a Senior Software Engineer with 11 years of experience building cross-platform client and server systems, currently driving high-impact features in SharePoint at Microsoft from Redmond. He blends mobile (Android/iOS), C/C++, and cloud expertise to deliver reliable, high-performance solutions—most notably redesigning SharePoint Copy/Move to use Azure Blobs and improving a 10GB benchmark by 70x while raising item reliability to 99.5%. Lei has deep pragmatic knowledge of backward compatibility and low-level concurrency/memory pitfalls, and has contributed performance-focused backend work to the widely used ONNX Runtime project. A Unix enthusiast and full‑stack practitioner, he’s equally comfortable authoring telemetry and monitoring frameworks as he is optimizing stream and memory management in ML runtimes.
11 years of coding experience
8 years of employment as a software developer
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at Nanjing UniversityMaster's degree, Computer Science, Master's degree, Computer Science at Peking University
Github Skills (11)
cuda10
roc10
c-language10
onnx10
cprogramming-language10
performance-optimization10
hardware-acceleration10
multithreading9
deep-learning8
ai-framework8
machine-learning8
Programming languages (6)
Github contributions (5)
microsoft/onnxruntime
Jul 2022 - Jan 2023
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:
Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:149 reviews, 54 commits, 156 PRs in 6 months
Contributions summary:Lei primarily contributed to the optimization of the ONNX Runtime, focusing on performance improvements related to stream handling and memory management. Their work involved modifying core components like `InferenceSession`, `RocmStream`, and `CudaStream` to enhance the efficiency of data transfer and resource allocation. Key changes included improving stream usage, and refactoring memory allocation to improve the efficiency and prevent thread safe issues. The contributions suggest an understanding of the underlying infrastructure to improve efficiency.
runtimetrainingtensorflowai-frameworkaccelerator
jslhcl/jslhcl.github.io
Feb 2015 - Jan 2025
Contributions:101 pushes in 10 years
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.