Protonu Basu
Software Engineer, Automated Parallel Programming
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolProtonu Basu is a software engineer specializing in automated parallel programming and deep learning compilers, currently advancing compiler and framework work at NVIDIA after five years as a research scientist at Meta. He brings eight years of industry experience plus a Ph.D. in Computer Science from the University of Utah and a long track record in parallel computing, polyhedral compilers, and performance-oriented code generation. Earlier roles include postdoctoral research at Lawrence Berkeley National Lab and systems-level development at Adobe, giving him fluency across research, production compilers, and library engineering. Protonu is comfortable moving between cutting-edge research and scalable product implementation, turning complex program analyses into performant runtime artifacts. Based in New York, he combines academic rigor with hands-on engineering to optimize large-scale ML workloads. An uncommon strength is his sustained focus on compiler-driven parallelism that bridges theoretical optimizations with real-world deep learning systems.
8 years of coding experience
12 years of employment as a software developer
The University of UtahBachelor’s Degree, Computer Science and Engineering, Bachelor’s Degree, Computer Science and Engineering at Indian Institute of Technology (Banaras Hindu University), Varanasi
English, Bengali, Hindi
Github Skills (24)
matrix10
python10
matrix-multiplication10
acceleration10
machine-learning10
multiplication10
facebook10
numpy10
deep-learning10
tensorflow10
gpu10
autograd10
neural-network10
tensor10
gpu-acceleration10
Programming languages (2)
Github contributions (5)
protonu/pytorch
Nov 2019 - Sep 2021
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Contributions:9 PRs, 164 pushes, 36 branches in 1 year 10 months
pythongpu-accelerationdeep-learninggpuacceleration
protonu/TensorComprehensions
May 2018 - Jun 2018
A domain specific language to express machine learning workloads.
Contributions:27 pushes, 6 branches in 15 days
workloadsdeep-learningexpressmachine-learningdomain
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.