Benoit Steiner
Technical Staff at Facebook
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolBenoit Steiner is a Technical Staff engineer in Mountain View with 12 years of experience building high-performance compilers and machine learning infrastructure at companies including Anthropic, Meta (Facebook), and Google. He blends deep backend systems expertise—working on XLA buffer management and Halide ONNX compilation—with pragmatic frontend improvements, having contributed UI enhancements to TensorFlow's TensorBoard. His career spans hands-on engineering and technical leadership across industry and research-focused teams, with prior roles in software architecture and product development dating back to the early 2000s. Benoit’s open-source footprint shows comfort across the stack, from numerical libraries like Eigen to ML tooling and docs, reflecting both low-level algorithmic thinking and attention to developer experience. Based in the heart of Silicon Valley, he pairs rigorous academic training from École Centrale and MINES ParisTech with a track record of shipping production-grade optimizations that improve both performance and usability. An often-overlooked strength is his ability to move fluidly between improving developer documentation and implementing compiler-level buffer optimizations, bridging gaps between users and low-level systems.
12 years of coding experience
18 years of employment as a software developer
Master's degree, Master's degree at Ecole centrale de ParisMaster's degree, Master's degree at MINES ParisTechHoche
Stackoverflow
1,469reputation
81kreached
37answers
0questions
eigen
top-5%
tensorflow
top-5%
Github Skills (42)
markdown10
javascript10
c-language10
matrix10
operation10
halide10
typescript10
onnx10
compiler-compiler10
markdown-it10
front-end-development10
typescript-types10
hla10
html10
compiler10
Programming languages (7)
Github contributions (5)
tensorflow/docs
Feb 2017 - Mar 2018
TensorFlow documentation
Role in this project:
Technical Writer
Technical WriterContributions:34 commits in 1 year 1 month
Contributions summary:Benoit primarily contributed to the project by modifying and enhancing the documentation. Their changes include improvements to the docstrings, adding compatibility notes, and updating reference links. Additionally, the user worked on generating Markdown documentation for various project components, including functions, classes, and modules. These efforts aimed to improve the clarity and completeness of the TensorFlow documentation.
deep-learningmachine-learningdeep-neural-networkstensorflow-tutorialstensorflow
halide/Halide
Nov 2018 - Feb 2020
a language for fast, portable data-parallel computation
Role in this project:Back-end Developer
Back-end DeveloperContributions:1 review, 122 commits, 58 PRs in 1 year 3 months
Contributions summary:Benoit primarily contributed to the Halide compiler by adding features related to compiling and converting ONNX models into Halide pipelines. Their contributions involved adding support for new ONNX operators, such as the Winograd convolution and split operations. They also improved existing features by implementing the use of symbolic shapes and making optimizations, improving efficiency. The user’s work demonstrates a focus on extending the Halide compiler's capabilities for deep learning model execution.
computationhexagonhalideparallelgpu
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.