Nirvedh Meshram
Principal Member Of Technical Staff at AMD
Madison, Wisconsin, United States
Summary
🤩
Rockstar🎓
Top SchoolNirvedh Meshram is a Principal Member of Technical Staff at AMD with a Ph.D. in Electrical and Computer Engineering and about four years of professional experience bridging academic research and production ML compiler engineering. He previously led medical imaging research as a postdoc at Columbia, developing pulse wave imaging and quantitative algorithms for carotid analysis, and brings deep expertise in image registration, segmentation, and high-performance scientific code. At Nod.ai (acquired by AMD) and now at AMD he focused on lowering ML frameworks to MLIR and advancing LLVM/LLVMGPU backends, contributing upstream to high-profile projects like the LLVM project and IREE. His open-source work includes MLIR legalizations, tensor vectorization, and GPU-specific optimizations for CUDA/ROCm and tensorcore pipelines, reflecting both compiler-level rigor and practical runtime concerns. Based in Madison, WI, he combines research-grade algorithmic thinking with hands-on systems engineering to turn complex numerical methods into efficient, production-ready compiler passes and runtime features. A less obvious strength is his ability to move between signal-processing problems in biomedical imaging and low-level compiler internals, enabling cross-domain solutions that optimize both accuracy and performance.
4 years of coding experience
8 years of employment as a software developer
Doctor of Philosophy (Ph.D.), Electrical and Computer Engineering, Doctor of Philosophy (Ph.D.), Electrical and Computer Engineering at University of Wisconsin-MadisonBachelor of Technology (B.Tech.), Bachelor of Technology (B.Tech.) at Veermata Jijabai Technological Institute (VJTI)
Hindi, Marathi
Stackoverflow
449reputation
12kreached
32answers
2questions
Github Skills (24)
c-language10
operation10
llvm10
tensorrt10
mlr10
roc10
compiler-design10
tensorflow10
cuda10
tensor10
cprogramming-language10
vectorization9
machine-learning9
compiler9
gpu-programming8
Programming languages (6)
Github contributions (5)
llvm/llvm-project
Jan 2022 - Apr 2025
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:36 reviews, 27 PRs, 23 pushes in 3 years 2 months
Contributions summary:Nirvedh's commits primarily focus on modifying and enhancing the MLIR (Multi-Level Intermediate Representation) compiler infrastructure within the LLVM project. They implemented legalizations for GPU operations, added support for vectorization of tensor operations, and addressed issues related to convolution and i1 type support. Furthermore, the user worked on improving the tensor padding calculations and fixed typos within the comments. Their work included making modifications to the MLIR code and adding new tests.
compilerstechnologiesclangsubmittoolchain
iree-org/iree
Feb 2022 - Oct 2022
A retargetable MLIR-based machine learning compiler and runtime toolkit.
Role in this project:Back-end Developer
Back-end DeveloperContributions:168 reviews, 19 commits, 173 PRs in 8 months
Contributions summary:Nirvedh contributed to the `iree-org/iree` repository, a machine learning compiler and runtime toolkit. Their work focused on implementing and optimizing various components related to the IREE compiler's LLVMGPU backend and related modules. The user implemented optimizations for the tensorcore pipeline and added support for dynamic shared memory and the appropriate attribute for workgroup local memory. The user also contributed to CUDA and ROCM related functionality.
mlirspirvvulkantensorflowcompiler
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.