Umang Yadav
Staff Software Engineer at AMD
Old Toronto, Ontario, Canada
Summary
🤩
Rockstar🎓
Top SchoolUmang Yadav is a Staff Software Engineer at AMD with eight years of experience building ML acceleration stacks, currently driving MLIR-based kernel generation work on rocMLIR and contributing to the ROCm ecosystem. He has deep compiler and graph-compiler expertise from leading roles on MIGraphX, UIF, and TVM-based toolchains, and earned an AMD spotlight award for his FP8 work. Prior experience includes building Relay/TVM graph transformers and high-performance kernels for ASICs at Untether AI and TVM-based compiler contributions at Huawei, showing a consistent focus on turning research-grade compiler ideas into production-ready inference tooling. Based in Old Toronto, he is an active open-source contributor with a public GitHub profile centered on ML acceleration projects, combining strong systems-level engineering with hands-on kernel and codegen optimization experience.
8 years of coding experience
8 years of employment as a software developer
High School Certificate Physics Chemistry Maths, High School Certificate Physics Chemistry Maths at Smt. V. D. Desai Wadiwala (Bhulka Bhavan) High SchoolBachelor of Technology (B.Tech.) Electronics and Communications Engineering, Bachelor of Technology (B.Tech.) Electronics and Communications Engineering at National Institute of Technology SuratMaster of Applied Science (MASc) Electrical and Computer Engineering, Master of Applied Science (MASc) Electrical and Computer Engineering at University of Toronto
English, Gujarati, Hindi
Github Skills (51)
opencl10
tvm10
vulkan10
javascript10
modular10
python10
llvm10
gpu-programming10
acceleration10
graph10
machine-learning10
cluster-computing10
cpu10
amd10
numpy10
Programming languages (9)
Github contributions (5)
umangyadav/AMDMIGraphX
Jun 2021 - Apr 2023
AMD's graph optimization engine.
Contributions:97 pushes, 12 branches in 1 year 9 months
amdoptimizationgraphgraph-optimization
umangyadav/tvm
Aug 2019 - Apr 2022
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Contributions:37 pushes, 11 branches in 2 years 8 months
cpugpu-programminggpu-accelerationtvmdeep-learning
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.