Ayan Moitra
Staff Software Engineer Tech Lead, Deep Learning Compilers at NVIDIA
California, United States
Summary
🤩
Rockstar🎓
Top SchoolAyan Moitra is a Staff Software Engineer and Tech Lead specializing in deep learning compilers, with seven years of industry experience building high-performance GPU-accelerated tooling for AI workloads. Currently leading XLA development at NVIDIA, he has a strong track record integrating cuDNN and CUDA optimizations into major open-source projects like TensorFlow/XLA and nGraph to accelerate fused MHA, convolutions, and matrix kernels. His background includes architecting GPU inference and training engines at MathWorks and backend optimizations for Nervana/Intel, giving him rare end-to-end expertise across compiler, runtime, and kernel implementation layers. He holds a Ph.D. in Computational Dynamics, where he developed novel GPU-parallel algorithms for nonlinear wave instabilities—a research-to-production trajectory that surfaces in his meticulous unit-testing and performance-focused contributions. Based in California, he combines deep numerical modeling instincts with pragmatic engineering to squeeze performance out of modern accelerators.
7 years of coding experience
11 years of employment as a software developer
The University of Maryland, College ParkMaster of Science - MS, Mechanical Engineering, Master of Science - MS, Mechanical Engineering at University of MarylandBachelor of Technology (B.Tech.), Bachelor of Technology (B.Tech.) at Indian Institute of Technology, Kharagpur
Bengali, Hindi, English
Github Skills (15)
cuda10
convolution10
compiler10
gpu-programming10
xla10
machine-learning10
c-language10
deep-learning10
tensorflow10
cprogramming-language10
cudnn10
python10
compiler-development9
unit-testing9
distributed-training7
Programming languages (3)
Github contributions (5)
DeepRec-AI/DeepRec
Jul 2019 - Mar 2021
DeepRec is a high-performance recommendation deep learning framework based on TensorFlow. It is hosted in incubation in LF AI & Data Foundation.
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:69 commits in 1 year 8 months
Contributions summary:Ayan contributed to the deep learning framework by implementing and fixing issues related to convolution operations, specifically addressing backward input and filter convolutions. The commits involve modifications to the CUDNN convolution rewriter, and depthwise convolution converters. The user also incorporated comments and addressed potential issues within the source code.
pythondeep-learningrecommendationdistributed-trainingmachine-learning
openxla/xla
Jul 2019 - Aug 2022
A machine learning compiler for GPUs, CPUs, and ML accelerators
Role in this project:Back-end Developer
Back-end DeveloperContributions:1 review, 54 commits, 6 PRs in 3 years 2 months
Contributions summary:Ayan primarily contributed to the XLA compiler, focusing on the integration of cuDNN APIs for grouped convolution operations. Their work included enabling the use of cuDNN backprop APIs for grouped convolutions, as well as handling depthwise forward and backward filter convolutions using cuDNN. They also addressed comments and resolved conflicts in the codebase, further refining and enhancing the compiler's functionality.
compilercommunity-drivenmachine-learningmodular
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.