Vidush Singhal
PhD Candidate
West Lafayette, Indiana, United States
Summary
🤩
RockstarVidush Singhal is a PhD candidate at Purdue University with six years of software engineering and machine learning systems experience, focused on bridging PyTorch and compiler infrastructures. He contributes to the high-profile torch-mlir project, implementing end-to-end tests and MLIR conversion patterns for ATen operations like boolean/logical ops and embedding bag, demonstrating deep familiarity with PyTorch internals and compiler shape functions. His work blends research rigor with practical engineering, validating transformations through Python test suites and targeted pattern modifications. Based in West Lafayette, Indiana, he brings academic depth to open-source engineering and reliable testing practices. Colleagues rely on him for translating complex model semantics into robust compilation behavior that surfaces correctness early.
5 years of coding experience
Github Skills (8)
compiler10
pytorch10
compiler-compiler10
python10
mlr10
testing9
machine-learning8
mlops7
Programming languages (6)
Github contributions (5)
llvm/torch-mlir
Jun 2022 - Aug 2022
The Torch-MLIR project aims to provide first class support from the PyTorch ecosystem to the MLIR ecosystem.
Role in this project:
ML Engineer
ML EngineerContributions:48 reviews, 7 commits, 14 PRs in 2 months
Contributions summary:Vidush's primary contribution involves adding end-to-end (E2E) testing support for various PyTorch operations (Aten functions) within the Torch-MLIR project. They implemented and tested functionalities for boolean operations, logical operations, and the embedding bag operation. This includes modifying the MLIR conversion patterns, adding test cases in Python, and incorporating shape function support to validate correct transformations. The user demonstrates a strong understanding of PyTorch's ATen operations and the MLIR compilation process within the context of the project.
pytorchmlirtorchcompilerecosystem
vid-999/torch-mlir
May 2022 - Aug 2022
The Torch-MLIR project aims to provide first class support from the PyTorch ecosystem to the MLIR ecosystem.
Contributions:100 pushes, 9 branches in 2 months
pytorchmlirtorchecosystemaims
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.