Tanay Varshney
Principal Engineer (IC6) at NVIDIA
California, United States
Summary
🤩
Rockstar🎓
Top SchoolTanay Varshney is a Principal Engineer at NVIDIA with 11 years of hands-on experience building and prototyping end-to-end vision systems for autonomous vehicles, aerial robotics, and remote sensing. He blends deep computer vision and SLAM expertise—working with point clouds, spectral and satellite/drone imagery—with practical skills in high-throughput data ingestion, time-series analysis, and data visualization to deliver actionable business insights. At NVIDIA he progressed rapidly through engineering levels, contributing to model deployment and inference tooling (including Triton examples and Torch-TensorRT docs) that streamline production ML workflows. Research-oriented and product-minded, he bridges academic rigor from his MS in Computer Science at NYU with production deployment know-how, making him effective across R&D and operational environments. An unsung strength is his emphasis on clear technical documentation and developer experience, improving adoption of complex GPU inference stacks.
11 years of coding experience
7 years of employment as a software developer
Bachelor of Technology (BTech), Computer Engineering, Bachelor of Technology (BTech), Computer Engineering at KJ Somaiya College of Engineering, VidyaviharMaster of Science - MS, Computer Science, Master of Science - MS, Computer Science at New York University
English, Hindi
Github Skills (18)
transformers10
continuous-deployment10
pytorch10
learn-ruby-on-rails10
python10
tensorrt10
user-manual10
machine-learning10
onnx10
basics10
ml-deployment10
triton10
tensorflow10
triton-server10
documentation10
Programming languages (3)
Github contributions (5)
triton-inference-server/tutorials
Dec 2022 - Mar 2023
This repository contains tutorials and examples for Triton Inference Server
Role in this project:
ML Engineer
ML EngineerContributions:8 reviews, 18 commits, 17 PRs in 3 months
Contributions summary:Tanay primarily contributed to the development and integration of machine learning models within the Triton Inference Server environment. Their work involved creating Python-based models for tasks such as image generation using Stable Diffusion, and deploying various models for quick deployment. They showcased expertise by including client-side scripts for interacting with the deployed models, indicating a focus on the end-to-end inference workflow. The user's contributions also included examples using PyTorch, TensorFlow and ONNX models.
pytorch/TensorRT
Jan 2022 - Aug 2022
PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT
Role in this project:Technical Writer
Technical WriterContributions:5 reviews, 22 commits, 14 PRs in 6 months
Contributions summary:Tanay's commits primarily focus on improving and updating the documentation for the `pytorch/tensorrt` repository. The changes involve fixing broken links in tutorials and updating API references, indicating a focus on enhancing the clarity and usability of the documentation. These contributions directly improve the learning experience for users of the Torch-TensorRT library.
cudapytorchjetsonnvidiadeep-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.