Subhankar Shah
Software Engineer at Google
Bengaluru, Karnataka, India
Summary
👤
Senior🎓
Top SchoolSubhankar Shah is a software engineer with 8 years of experience, currently at Google in Bengaluru, focused on backend systems and ML compiler work. He has a strong track record in open-source, contributing to TensorFlow and OpenXLA’s StableHLO—helping implement operations like MulOp and enabling StableHLO-to-MHLO conversions critical to TensorFlow’s transition. Prior roles span startups and scale-ups including a technical lead position at Interview Kickstart, a co-founder stint, and backend engineering at Cure.Fit, demonstrating both product and systems chops. He enjoys competitive programming and keeps sharp by participating in contests, while avidly reading about system architecture and AI. Known for pragmatic engineering, he blends deep technical contributions to ML infrastructure with hands-on delivery in production environments.
8 years of coding experience
4 years of employment as a software developer
Bachelor of Technology, Computer Science, Bachelor of Technology, Computer Science at Indian Institute of Technology (Banaras Hindu University), Varanasi
Github Skills (12)
tensorrt10
machine-learning10
tensor10
tensorflow10
stable10
operation10
mlr10
python9
deeplearning-ai9
deep-learning9
compiler9
deep-neural-networks8
Programming languages (5)
Github contributions (5)
openxla/stablehlo
Oct 2022 - Jan 2023
Backward compatible ML compute opset inspired by HLO/MHLO
Role in this project:
Back-end Developer
Back-end DeveloperContributions:351 reviews, 38 commits, 73 PRs in 3 months
Contributions summary:Subhankar primarily contributed to the implementation and testing of the `MulOp` operation within the StableHLO framework. Their work involved adding specifications, interpreters, and tests for various data types, including integers, floating-point numbers, and complex numbers. The user also made changes to other operations, such as CollectivePermuteOp and SetDimensionSizeOp, and added the PartitionIdOp and support for bounds in the sort operation. This demonstrates a focus on expanding the functionality and improving the testing of the StableHLO library.
hlomachine-learningcompute
tensorflow/tensorflow
Jan 2023 - Jan 2023
An Open Source Machine Learning Framework for Everyone
Role in this project:Back-end Developer
Back-end DeveloperContributions:2 commits in 1 day
Contributions summary:Subhankar contributed to the StableHLO to MHLO conversion within the tensorflow/tensorflow repository, focusing on inferring bounds for return types in the AllToAll and SortOp operations. These changes involved modifying the MHLO dialect code to handle StableHLO conversions. The user also removed test patterns and integrated StableHLO updates and fixes, ensuring the compatibility of the StableHLO to MHLO transition. This work is crucial for the ongoing transition of the TensorFlow project.
pythondata-sciencedeep-learningmlmachine-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.