Tejash Shah
Product Manager, AI SW Platform
San Diego, California, United States
Summary
🤩
Rockstar🎓
Top SchoolTejash Shah is a product-minded AI software leader with 7 years in AI/ML platform and a long engineering pedigree spanning Qualcomm, AMD and NVIDIA. He has driven compiler and runtime integrations for edge and accelerator hardware—building Snapdragon Adreno/NPU support in TVM and contributing 3D batch-norm support to AMD's MIOpen—bridging low-level kernel work with product roadmaps. At OctoAI/NVIDIA he led cross-geo teams to expand ML compiler support for customers like Samsung and ByteDance, and he mentors engineers on kernel-level optimization for transformer workloads. He combines hands-on GPU/HIP/OpenCL performance tuning with strategic partner engagement and a knack for translating hardware nuances into developer-friendly tooling. Based in San Diego, he pairs a master’s in CS with deep domain experience in inference and training stacks, often surfacing pragmatic trade-offs that keep both performance and ecosystem adoption aligned.
7 years of coding experience
18 years of employment as a software developer
Master's Degree Computer Science, Master's Degree Computer Science at The University of Texas at DallasBachelor's Degree Computer Science, Bachelor's Degree Computer Science at Gujarat University
English, Hindi, Gujarati
Github Skills (6)
machine-learning10
c-language10
cprogramming-language10
batch-normalization10
opencl9
unit-testing9
Programming languages (4)
Github contributions (5)
ROCm/MIOpen
Aug 2018 - Jul 2020
AMD's Machine Intelligence Library
Role in this project:
Back-end Developer
Back-end DeveloperContributions:3 reviews, 98 commits, 8 PRs in 1 year 11 months
Contributions summary:Tejash focused on adding support for 3D batch normalization to AMD's Machine Intelligence Library (MIOPEN). Their work included implementing the necessary driver functionalities for 3D batchnorm, modifying existing API calls for 3D tensor descriptors, and adding unit tests for the new 3D batchnorm functionalities. The user's contributions were focused on adding support for various 3D tensor processing in the batch normalization APIs.
amdcluster-computingmachine-learningblasmachine-intelligence
TejashShah/HuggingFace
Apr 2023 - Apr 2023
Contributions:3 pushes, 1 branch in 4 days
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.