Sam Parker
Product Lead at Meta
San Francisco, California, United States
Summary
🤩
RockstarSam Parker is a Product Lead based in San Francisco who builds experiences where art meets data and systems, focused on what people actually feel and do. With three years of product and engineering experience spanning media and tech, B2C and B2B, and both 0-to-1 and scale phases, he navigates core experiences and platform work fluidly. At Meta he leads cross-functional teams to translate words, data, and systems into measurable product outcomes. Sam pairs hands-on engineering chops—evidenced by contributions to the high-profile PyTorch project, including inductor enhancements and dynamic-shape fixes—with product strategy, allowing him to bridge ML infra and user-facing features. He favors pragmatic solutions that respect both technical constraints and human context. Colleagues know him for clear priorities, an appetite for difficult trade-offs, and an uncommon fluency between product storytelling and compiler-level engineering.
3 years of coding experience
English, French
Github Skills (11)
pytorch10
machine-learning10
irr10
tensor10
gpu10
python10
intermediate-code10
autograd9
cprogramming-language9
neural-network9
c-language9
Programming languages (3)
Github contributions (5)
pytorch/pytorch
Jun 2023 - May 2026
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:679 reviews, 382 PRs, 1354 pushes in 2 years 11 months
Contributions summary:Sam primarily contributed to the PyTorch framework by developing and implementing features for the inductor, a deep learning compiler. Their contributions included adding a C++ implementation for random integer generation to the IR (Intermediate Representation) and providing support for using the 'stream' parameter in AOT (Ahead of Time) mode. Further, they enhanced the framework by decomposing the aten.rrelu_with_noise operator. They also fixed an issue with dynamic shape handling within the inductor and enhanced the FX graph caching mechanism.
gpu-accelerationneural-networkpythonautogradgpu
masnesral/pytorch
Apr 2024 - May 2026
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Contributions:35 pushes, 27 branches in 2 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.