Nithin Meganathan
Member Of Technical Staff at AMD
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolNithin Meganathan is a robotics and systems engineer with 10 years of experience, currently a Member of Technical Staff at AMD focused on fast ML compilers and kernels. He holds a Master's in Robotic Systems Development from CMU and combines expertise in computer vision, mapping, and machine learning with hands-on hardware-aware systems work. Previously he built ML software and inference infrastructure at Nod Labs (acquired by AMD) and contributed to edge vision and aerial manipulator research projects. As an active back-end contributor to the IREE project, he implemented GPU-focused features—SPIR-V/Vulkan enhancements, Intel ARC and Pascal target triples, and early HIP HAL driver components—highlighting low-level runtime and cross-hardware portability skills. Based in San Francisco, he bridges research and product engineering to ship end-to-end robotic and ML systems. Colleagues describe him as a pragmatic problem-solver who enjoys taking compiler-level challenges down to the metal.
10 years of coding experience
3 years of employment as a software developer
Bachelor of Technology, Distinction, Mechanical Engineering, Bachelor of Technology, Distinction, Mechanical Engineering at Amrita Vishwa VidyapeethamHigh School, High School at Suguna PIP School - IndiaMaster of Science, Robotic Systems Development, Master of Science, Robotic Systems Development at Carnegie Mellon University School of Computer Science
English, Tamil
Github Skills (12)
spirv10
compiler10
compiler-compiler10
c-language10
cprogramming-language10
system-programming10
gpu10
runtimes9
mlr9
opencl9
vulkan9
cuda8
Programming languages (4)
Github contributions (5)
iree-org/iree
Sep 2021 - Feb 2025
A retargetable MLIR-based machine learning compiler and runtime toolkit.
Role in this project:
Back-end Developer / Systems Engineer
Back-end Developer / Systems EngineerContributions:1 release, 35 reviews, 71 PRs in 3 years 5 months
Contributions summary:Nithin contributed significantly to the IREE compiler and runtime toolkit. Their work involved enhancing the SPIR-V and Vulkan backends, specifically by handling GPU address spaces for OpenCL and adding target triple support for Intel ARC A770 and Pascal GPUs. Additionally, they initiated the implementation of a HIP (Heterogeneous Interface for Portability) HAL driver, introducing core components like the driver, buffer, and allocator implementations, moving towards the completion of a new AMD backend. This indicates a focus on low-level system interactions and hardware support.
mlirspirvvulkantensorflowcompiler
nithinsubbiah/torch-mlir
Jun 2022 - Mar 2024
The Torch-MLIR project aims to provide first class support from the PyTorch ecosystem to the MLIR ecosystem.
Contributions:103 pushes, 13 branches in 1 year 9 months
pytorchmlirtorchecosystemwandb
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.