Subhadeep Bhattacharya
Senior Engineer, Sys SW at NVIDIA
California, United States
Summary
🤩
Rockstar🎓
Top SchoolSubhadeep Bhattacharya is a Senior Engineer specializing in system software, high-performance computing, and distributed deep learning, currently driving system-level architecture at NVIDIA in California. With seven years of experience bridging academic research and industry—spanning microservices and full-stack work at Infosys to research assistantship at Florida State University—he focuses on performance-sensitive distributed systems and communication libraries. He has practical impact on flagship open-source projects like Open MPI, where he optimized memory management and reduced footprints in UCX/SPML components to boost collective operation performance. Comfortable moving between research and production, he brings a rare combination of deep systems knowledge and hands-on engineering that accelerates scalable ML and HPC stacks.
7 years of coding experience
10 years of employment as a software developer
Doctor of Philosophy - PhD (Drop-out) Computer Science, Doctor of Philosophy - PhD (Drop-out) Computer Science at Florida State UniversityHigh School, High School at Hooghly Collegiate SchoolBachelor of Technology (B.Tech.) Information Technology, Bachelor of Technology (B.Tech.) Information Technology at St. Thomas' College of engineering and technologyHigh School, High School at Magra Uttamachandra High School
English, Hindi, Bengali
Github Skills (13)
c1710
memory-management10
ucx10
mpi10
openmpi10
hpc10
c1110
performance-optimization10
fortran9
algorithms8
data-structures8
algorithm8
data-structure8
Programming languages (5)
Github contributions (5)
open-mpi/ompi
Jun 2021 - Aug 2021
Open MPI main development repository
Role in this project:
Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:15 commits, 7 PRs, 1 comment in 2 months
Contributions summary:Subhadeep primarily focused on optimizing the Open MPI (OMPI) library, a core component of the OpenSHMEM (OSHMEM) library, with a focus on memory management and performance. They implemented changes to reduce the memory footprint of the UCX SPML component by introducing dynamic caching mechanisms. Several commits address issues related to memory allocation, deallocation, and general memory usage within the OpenSHMEM runtime environment, directly impacting the performance of communicator creation and collective operations.
openmpichpcmpifortran
sb17v/ompi
Jun 2021 - Dec 2022
Open MPI main development repository
Contributions:15 reviews, 12 PRs, 81 pushes in 1 year 6 months
mpiswigopenmp
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.