Shiva Shahrokhi
Senior AI Engineer at Microsoft
Greater Seattle Area United States
Summary
🤩
Rockstar🎓
Top SchoolShiva Shahrokhi is a Senior AI Engineer based in the Greater Seattle Area with 11 years of experience building production ML systems and low-level compiler tooling. He holds a PhD in Electrical and Computer Engineering (Robotics and Control) and has moved from applied research roles into machine learning engineering at Google and now Microsoft. Shiva contributes to high-impact open-source projects like XLA, where his work on GPU code generation, fusion handling, and GEMM optimizations shows deep expertise in compiler internals and accelerator-aware performance engineering. He combines research rigor with product-focused delivery, having worked on applied research at Expedia and completed an intensive data science fellowship. Colleagues rely on him to bridge algorithmic innovation and systems implementation, especially when squeezing performance from GPUs and ML accelerators. An underappreciated strength is his knack for adding pragmatic C APIs and debugging hooks that make low-level systems far easier to integrate and operate.
11 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy - PhD Electrical and Computer Engineering Robotics and Control, Doctor of Philosophy - PhD Electrical and Computer Engineering Robotics and Control at University of HoustonBachelor of Engineering - BE Computer Software Engineering, Bachelor of Engineering - BE Computer Software Engineering at Iran University of Science and Technology
English, Persian, French
Github Skills (8)
xla10
compiler10
gpu-programming10
c-language10
cprogramming-language10
compile9
compilation9
cuda8
Programming languages (6)
Github contributions (5)
openxla/xla
Mar 2022 - Aug 2022
A machine learning compiler for GPUs, CPUs, and ML accelerators
Role in this project:
Back-end Developer
Back-end DeveloperContributions:17 commits in 4 months
Contributions summary:Shiva's contributions primarily involve modifying and improving the XLA compiler's GPU-related components. They've made changes to the `IrEmitterUnnested` class, including modifications to fusion and select/scatter operations, and the `gemm_thunk.cc` file for GEMM operations, demonstrating an understanding of compiler internals and GPU code generation. The changes include adding PJRT C API methods for device properties and debugging information. These edits suggest they are working on low-level optimizations and interfacing with other systems.
compilercommunity-drivenmachine-learningmodular
sshahrokhi/jax
May 2022 - Mar 2024
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
Contributions:47 pushes, 7 branches in 1 year 10 months
pytorchpythonjitgpunumpy
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.