Ryan Lai
Software Engineer II at Uber
Bellevue, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolRyan Lai is a Software Engineer II with 10 years of experience building production-grade systems for ML and large-scale services, currently contributing to Uber Maps' DeepETA team in Bellevue. He previously helped ship Windows Machine Learning performance tooling and bootstrapped DirectML as a PyTorch backend at Microsoft, creating an end-to-end performance testing system and instrumentation used for daily builds and releases. His work on the high-profile microsoft/Windows-Machine-Learning repo includes programmatic PIX capture, image-model input handling, and richer performance metadata—efforts that improved real-world model evaluation and debugging. Early roles at Amazon and Goldman Sachs show a pattern of rapidly delivering internal systems that improve operational efficiency. He holds a dual background in Computer Science and Linguistics from UIUC, which informs his analytical approach to systems and tooling.
10 years of coding experience
4 years of employment as a software developer
University of Illinois Urbana-Champaign
Github Skills (8)
windows10
machine-learning10
onnx10
performance-optimization10
directx9
cprogramming-language9
c-language9
tensorflow8
Programming languages (4)
Github contributions (5)
microsoft/Windows-Machine-Learning
Mar 2019 - Nov 2021
Samples and Tools for Windows ML.
Role in this project:
ML Engineer
ML EngineerContributions:12 releases, 37 reviews, 15 commits in 2 years 8 months
Contributions summary:Ryan focused on enhancing the Windows Machine Learning tools by implementing features related to performance analysis and image input handling. They added programmatic capture capabilities for the PIX tool, enabling performance measurements during model evaluation. Furthermore, they introduced support for models with image denotation, adding new input handling logic for image-based models and associated test files. In addition, they fixed return error codes, improved versioning, and added the ability to add metadata to performance output files.
machine-learningwindowsaimldeep-learning
ryanlai2/Multithreaded-Stock-Indicator-Generator
Oct 2016 - Jan 2017
Contributions:13 pushes, 1 branch in 3 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.