Mehdi Amini
Distinguished Engineer at NVIDIA
Switzerland
Summary
🤩
Rockstar🎓
Top SchoolMehdi Amini is a Distinguished Engineer with 11 years of experience building compilers, runtimes, and deep learning infrastructure, currently driving deep learning software and compiler efforts at NVIDIA from Switzerland. He led core MLIR and OpenXLA work at Google, bridging XLA, MLIR, and IREE to make compiler toolchains more modular and production-ready. His open-source footprint includes substantive backend contributions to TensorFlow, LLVM/Clang, Triton and the XLA runtime—work that often focused on refactoring, dependency isolation, and enabling scalable ThinLTO and MLIR-based pipelines. Comfortable at the intersection of research and production, he has repeatedly improved build systems, testing, and developer productivity for large codebases. Notably, he combines low-level compiler expertise with pragmatic MLOps thinking, enabling smoother transitions from compiler research to deployable ML runtimes.
11 years of coding experience
16 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Ecole nationale supérieure des Mines de ParisMaster's degree, Fundamental Research in Computer Science, Master's degree, Fundamental Research in Computer Science at University of Strasbourg
English, French
Github Skills (36)
debugging10
c-language10
operation10
clang-cl10
dwarf10
compiler-development10
llvm10
c-compiler10
machine-learning10
build-system10
flang10
mlr10
lldb10
triton10
compiler-design10
Programming languages (22)
Github contributions (5)
llvm/llvm-project
Sep 2021 - Jan 2023
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:2614 reviews, 9 commits, 565 PRs in 1 year 4 months
Contributions summary:Mehdi's commits focus on reverting code changes related to the I/O APIs in the Flang runtime for offload builds. The commits suggest a regression in Windows builds, specifically related to I/O operations. Additional commits include reverting changes associated with the complex log1p and fixing compiler crashes. Finally, the commits point to reverting several MLIR changes.
compilerllvmtoolchain
tensorflow/tensorflow
Dec 2018 - Jan 2023
An Open Source Machine Learning Framework for Everyone
Role in this project:Backend Developer
Backend DeveloperContributions:422 reviews, 898 commits, 17 PRs in 4 years 1 month
Contributions summary:Mehdi primarily contributed to the XLA (Accelerated Linear Algebra) component within the TensorFlow repository. Their work focused on refactoring and updating XLA to reduce dependencies, specifically modifying include paths to use `tsl/c/tsl_status.h` and migrating `bfloat16_lib` to TSL. They also updated XLA to use bfloat16 from TSL and fixed XLA test macros, and introduced XLA-specific testing configurations.
machine-learningtensorflowpythondeep-learningdeep-neural-networks
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.