Kushashwa Shrimali
Senior Member Technical at The D. E. Shaw Group
Bengaluru, Karnataka, India
Summary
🤩
Rockstar🎓
Top SchoolKushashwa Shrimali is a Senior Member Technical with a decade of engineering experience building and hardening Python infrastructure and ML tooling, currently focused on Python infra at D. E. Shaw after stints at Lightning AI and contributing to PyTorch development. A gold-medalist CSE graduate from IIIT Naya Raipur, he blends deep systems work in C++/CUDA and compiler practice with higher-level ML engineering in PyTorch and Lightning, including notable contributions to a Python compiler (lpython) and distributed training support (Horovod) in PyTorch Lightning. He has shipped production-facing features and tests across open-source AI stacks—fixing parsers, enabling type promotion in tensor iterators, and improving image-embedding and QA pipelines—demonstrating both research depth and pragmatic engineering. Equally comfortable in Rust and low-level CUDA as in Python ML libraries, he has a track record of improving developer-facing libraries and reproducible training workflows. Outside core contributions, he writes technical tutorials and maintains a public blog and YouTube channel, signaling a commitment to knowledge sharing and community impact.
9 years of coding experience
5 years of employment as a software developer
Bachelor’s Degree Computer Science, Bachelor’s Degree Computer Science at International Institute of Information Technology Naya RaipurHigh School CBSE - Science Stream, High School CBSE - Science Stream at Kendriya Vidayalaya STPS
Hindi, English
Stackoverflow
193reputation
5kreached
0answers
3questions
Github Skills (30)
pytorch10
opencv10
parser10
distributed-training10
python10
scikit10
testing10
machine-learning10
svm10
pytorch-lightning10
horovod10
compiler-compiler10
deep-learning10
parsing10
parse10
Programming languages (16)
Github contributions (5)
Lightning-Universe/lightning-flash
Feb 2022 - Nov 2022
Your PyTorch AI Factory - Flash enables you to easily configure and run complex AI recipes for over 15 tasks across 7 data domains
Role in this project:
ML Engineer
ML EngineerContributions:2 releases, 202 reviews, 112 commits in 8 months
Contributions summary:Kushashwa primarily contributes to the `lightning-flash` repository by fixing examples, modifying pretraining transforms, and propagating collate functions within the image embedding module. They also address issues related to dependencies and the use of external libraries like `nltk` and `vissl`. Their work seems focused on improving the functionality and integrating new features related to image embedding and question-answering tasks within the PyTorch Lightning framework.
torch-geometricicevisionconfigurepytorch-lightningdomains
Lightning-AI/pytorch-lightning
Feb 2022 - Nov 2022
Pretrain, finetune ANY AI model of ANY size on multiple GPUs, TPUs with zero code changes.
Role in this project:ML Engineer
ML EngineerContributions:4 releases, 223 reviews, 34 commits in 8 months
Contributions summary:Kushashwa contributed to the PyTorch Lightning library, specifically focusing on integrating and supporting Horovod for distributed training. Their work involved modifications to the testing suite, adding support for gradient accumulation with Horovod, and ensuring compatibility with various configurations including CPU and GPU training. They also refactored tests and addressed issues related to checkpointing, as well as providing some general code cleanup and preparation for the 1.6.0rc1 release.
pythonheadachespytorch-modelsdata-sciencehandling
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.