Gary Miguel
Member Of Technical Staff at Anthropic
United States
Summary
🤩
Rockstar🎓
Top SchoolGary Miguel is a Member of Technical Staff with 11 years of experience building developer tooling, ML interoperability, and scalable infrastructure across Google, Microsoft, Verily, and startups, now focused at Anthropic. He led platform and productivity teams at Google, ran ONNX converters as a principal engineer at Microsoft, and contributed to high-profile open standards like ONNX and ONNX Runtime to improve PyTorch interoperability and inference performance. Comfortable shifting between hands-on engineering and technical leadership, he has a track record of improving developer happiness and CI workflows while partnering with hardware vendors to accelerate models. He holds an MS in Computer Science from Stanford and a BS from UC Berkeley, and donates 10% of his earnings to effective charities, reflecting a values-driven approach to impact. An underappreciated strength is his knack for turning messy build/test systems into reliable developer-facing platforms that scale with growing teams and device types.
11 years of coding experience
15 years of employment as a software developer
Bachelor's degree Computer Science, Bachelor's degree Computer Science at University of California, BerkeleyMaster of Science - MS Computer Science, Master of Science - MS Computer Science at Stanford University
Github Skills (11)
pytorch10
machine-learning10
deep-learning10
onnx10
tensorflow10
python10
documentation9
hardware-acceleration9
neural-network9
scikit7
scikit-learn7
Programming languages (30)
Github contributions (5)
onnx/onnx
Apr 2021 - Jun 2022
Open standard for machine learning interoperability
Role in this project:
ML Engineer
ML EngineerContributions:27 reviews, 23 commits, 22 PRs in 1 year 2 months
Contributions summary:Gary primarily contributed to the ONNX project by making improvements to the documentation and fixing various compilation warnings. The commits include updating the documentation for the "Where" operator to reflect multidirectional broadcasting and correcting some bugs in the Resize implementation. Furthermore, the user addressed compilation warnings related to deprecated features and fallthrough statements. The user also worked on refactoring the code by converting Python comments to type annotations, enhancing code clarity.
pytorchmxnetdeep-learninginteroperabilitymachine-learning
microsoft/onnxruntime
Sep 2021 - Jul 2022
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:ML Engineer
ML EngineerContributions:84 reviews, 33 commits, 42 PRs in 10 months
Contributions summary:Gary primarily focused on integrating and testing PyTorch-related components within the ONNX Runtime framework. They updated the code to include `pytorch_export_contrib_ops` in inference builds and made changes to incorporate it into training sessions. Their contributions involved renaming and moving files, enabling tests within the CI pipeline, and modifying `orttrainer.py` to support the new components. These changes aimed to enhance the framework's interoperability with PyTorch and expand its capabilities.
runtimetrainingtensorflowai-frameworkaccelerator
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.