Arshi Bhatnagar
Member Of Technical Staff at OpenAI
San Jose, California, United States
Summary
🤩
Rockstar🎓
Top SchoolAnupam Bhatnagar is a software engineer specializing in AI/ML who blends a PhD-level mathematics background with six years of industry experience at Google, Meta, and Unity. He builds production-grade ML infrastructure and features—contributing to high-profile open-source projects such as PyTorch and FairScale—while improving CI/CD, testing, and performance tooling for large-scale training. Previously a tenure-track assistant professor, he brings rigorous mathematical insight to practical problems like optimizer hooks, memory profiling, and distributed training scalers. Based in Sunnyvale, he pairs academic depth with hands-on MLOps and QA expertise, often surfacing subtle performance and reproducibility improvements that benefit both researchers and engineers.
6 years of coding experience
5 years of employment as a software developer
Bachelor’s Degree Major: Computer Science Minor: Operations Research, Bachelor’s Degree Major: Computer Science Minor: Operations Research at Cornell UniversityHigh School, High School at Lynbrook High School
English, Hindi, Japanese
Github Skills (26)
pytorch10
github-ci10
python10
machine-learning10
cicd10
deeplearning-ai10
deep-learning10
gpu10
profiling10
dotnet-core10
autograd10
deep-reinforcement-learning10
tensor10
csharp10
githubaction-workflow10
Programming languages (6)
Github contributions (5)
facebookresearch/fairscale
Nov 2021 - Dec 2022
PyTorch extensions for high performance and large scale training.
Role in this project:
MLOps Engineer
MLOps EngineerContributions:1 release, 81 reviews, 27 commits in 1 year 1 month
Contributions summary:Arshi primarily focused on improving the build and release process and incorporating testing within the repository. They set up a pre-commit GitHub Action to enforce code style and linting, updated Python versions, and refactored CI configurations to use GitHub Actions. Additionally, they made changes to enable and test features related to the ShardedGradScaler.
pytorchdeep-learningmachine-learningscaletraining
Unity-Technologies/ml-agents
Sep 2019 - Aug 2020
The Unity Machine Learning Agents Toolkit (ML-Agents) is an open-source project that enables games and simulations to serve as environments for training intelligent agents using deep reinforcement learning and imitation learning.
Role in this project:QA Engineer / Test Automation Engineer
QA Engineer / Test Automation EngineerContributions:115 commits in 11 months
Contributions summary:Arshi's commits primarily involve modifications to test files, specifically within the `ML-Agents` project. The changes include the creation and modification of edit mode tests, and testing various Academy and Agent functions. These tests appear to validate the core functionality and initialization procedures of the ML-Agents framework. They're focused on ensuring the correct behaviour of the agents, academy and brains within the Unity environment.
servedeep-reinforcement-learningtrainingunityreinforcement
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.