Max Ren
Member Of Technical Staff at Anthropic
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolMax Ren is a software engineer with 7 years of experience specializing in on-device ML inference and performance optimization, currently a Member of Technical Staff at Anthropic after building end-to-end on-device solutions at Meta. He contributed to PyTorch core—most notably CoreML integration, quantization, and mobile profiler tooling—helping enable efficient deployment on Apple devices and deeper memory/event introspection. Based in San Francisco and trained at Cornell, Max blends systems-level engineering with ML deployment expertise to ship production-ready tooling that bridges research models and constrained hardware. A detail-oriented problem solver, he has fixed subtle platform bugs (e.g., CoreML GPU/CPU execution flags) and added APIs that make profiling and debugging mobile ML more practical.
8 years of coding experience
4 years of employment as a software developer
Bachelor of Science - BS Computer Science, Bachelor of Science - BS Computer Science at Cornell University
Chinese
Github Skills (19)
pytorch10
machine-learning10
coreml10
quantization10
ios9
performance-analytics9
performance-monitor9
performance-measurement9
performance-monitoring9
gpu9
performance-analysis9
performance-tuning9
profiling8
neural-network7
python7
Programming languages (4)
Github contributions (5)
pytorch/pytorch
Sep 2021 - Nov 2022
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
ML Engineer
ML EngineerContributions:66 reviews, 160 commits, 110 PRs in 1 year 2 months
Contributions summary:Max primarily contributed to the CoreML integration within the PyTorch framework, focusing on quantization and performance optimization for deployment on Apple devices. Their work involved adding and modifying pre-processing steps to enable CoreML model quantization. The user implemented new APIs for profiling backend memory events and added targets for testing mobile profilers, thus enabling performance and memory usage analysis. The user also addressed a CoreML GPU flag issue that prevented CPU execution and incorporated a special throw macro for CoreML API errors.
gpu-accelerationneural-networkpythonautogradgpu
mcr229/XNNPACK
Dec 2023 - Aug 2025
High-efficiency floating-point neural network inference operators for mobile, server, and Web
Contributions:97 pushes, 32 branches in 1 year 8 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.