Phaneesh Barwaria
ML Compiler Engineer at nod.ai
Delhi, India
Summary
🤩
Rockstar🎓
Top SchoolPhaneesh Barwaria is an ML compiler engineer with 7 years of experience building performant machine learning tooling and deployment infrastructure from Delhi, India. He currently contributes to nod.ai, focusing on compiler integrations and runtime improvements, and previously held engineering roles at OnePlus where he moved from software engineer to senior engineer. His open-source work includes enhancing SHARK Studio’s web UI and Vulkan/device targeting for SHARK+IREE, reflecting practical expertise in ML runtimes, cross-platform GPU support, and test automation. With an academic foundation from IIT Delhi and a BE in Computer Science, he combines research-rooted rigor with hands-on DevOps and engineering pragmatism. Notably, he has tackled platform-specific issues like Apple M-series target triples and automatic Vulkan device selection—small fixes that materially improve reproducibility and user experience.
7 years of coding experience
2 years of employment as a software developer
Bachelor of Engineering - BE, Computer Science, Bachelor of Engineering - BE, Computer Science at Vidyalankar Institute of Technology, MumbaiIndian Institute of Technology Delhi (IIT Delhi)
Github Skills (10)
vulkan10
machine-learning10
pytest10
python10
mlr9
deep-learning9
pytorch9
nvidia8
amd8
cicd8
Programming languages (3)
Github contributions (5)
nod-ai/SHARK-Studio
Aug 2022 - Jan 2023
SHARK Studio -- Web UI for SHARK+IREE High Performance Machine Learning Distribution
Role in this project:
ML Engineer & DevOps Engineer
ML Engineer & DevOps EngineerContributions:96 reviews, 38 commits, 147 PRs in 5 months
Contributions summary:Phaneesh primarily focused on improving the SHARK Studio web UI for the SHARK+IREE machine learning distribution. Their contributions include identifying and adjusting target triples for Apple M2 devices and marking Vulkan tests as "xfail" for M1 failures. They simplified the testing interface, added support for choosing a Vulkan device, and integrated a vulkan target env. Furthermore, they added the ability to add cli arguments and automatic selection of vulkan device.
pytorchcudaamdheterogeneousdeep-learning
PhaneeshB/torch-mlir
Jul 2022 - Oct 2024
The Torch-MLIR project aims to provide first class support from the PyTorch ecosystem to the MLIR ecosystem.
Contributions:63 pushes, 14 branches in 2 years 2 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.