Krish Mehta
Senior Verification Engineer
California, United States
Summary
🤩
Rockstar🎓
Top SchoolKrish Mehta is a verification engineer with 7 years of experience specializing in CPU memory subsystem and microarchitectural validation, currently driving A0-ready RTL quality at NVIDIA. He combines deep RTL debug, coverage closure, and testbench performance optimization—having improved testbench cycles-per-second by 20–50%—with collaborative design and architecture verification. Krish has extended his impact into software-scale testing and backend tooling through meaningful contributions to the widely used Triton Inference Server, improving its client libraries, Python backend APIs, and reliability tests for large payloads and memory growth. A UC San Diego MS graduate and IIT Bombay alumnus, he pairs strong academic grounding with hands-on ASIC and verification delivery across multiple NVIDIA roles. Outside work he channels system-level curiosity into building custom mechanical keyboards and a sustained passion for games and computers.
7 years of coding experience
6 years of employment as a software developer
University of California, San DiegoIndian Institute of Technology Bombay
Github Skills (21)
client-library10
c-language10
back-end-development10
apidoc10
python10
testing10
http10
inference10
java10
javas10
triton10
api10
triton-server10
backend10
cprogramming-language10
Programming languages (5)
Github contributions (5)
triton-inference-server/python_backend
Jun 2022 - Jan 2023
Triton backend that enables pre-process, post-processing and other logic to be implemented in Python.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:251 reviews, 18 commits, 84 PRs in 7 months
Contributions summary:Krish contributed to the Python backend of a Triton Inference Server. They implemented an API for Python models to provide configuration details, including max batch size, inputs, and outputs, enhancing auto-completion functionality. The user also addressed code reviews, fixed bugs, and refactored code, improving the overall stability and maintainability of the backend. Key modifications included adding an output buffer query API and fixing handling for empty GPU tensors.
post-processingpre-processpythonbackendpython3
triton-inference-server/server
Jul 2021 - Jan 2023
The Triton Inference Server provides an optimized cloud and edge inferencing solution.
Role in this project:Backend Engineer
Backend EngineerContributions:1 release, 370 reviews, 55 commits in 1 year 6 months
Contributions summary:Krish primarily focused on modifying and enhancing test cases related to the Triton Inference Server. Their contributions included updating tests for GRPC errors, specifically addressing large payload issues. They also added and modified weekly tests for memory growth and improved the general testing framework by incorporating test duration data. These modifications are focused on improving reliability and performance testing for the backend.
nvidia-dockernvidiadeep-learninggpuinference
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.