Rahulkumar Gayatri
Pleasanton, California, United States
Summary
🤩
Rockstar🎓
Top SchoolRahulkumar Gayatri is an application performance specialist with 10 years of experience in high-performance computing and exascale readiness, currently working in the Programming Environments and Models group at NERSC, LBNL. He brings deep research credentials—a PhD focused on multicore task-based models and software transactional memory—and practical production experience tuning parallel algorithms across CUDA and OpenMPTarget backends. His notable open-source contributions to the widely used Kokkos ecosystem include implementing hierarchical parallelism, reductions, deep-copy optimizations, and CMake/testing improvements that directly impact performance portability. Comfortable bridging research and engineering, he has moved from academic postdoctoral work at Berkeley Lab to hands-on performance engineering, often tackling low-level atomic and memory abstractions that others avoid. Based in Pleasanton, CA, he combines rigorous mathematical training with systems-level coding to make applications run efficiently on next-generation supercomputers.
10 years of coding experience
2 years of employment as a software developer
Bachelor's degree, Honors in Mathematics, 3.95/5.00, Bachelor's degree, Honors in Mathematics, 3.95/5.00 at Sri Sathya Sai Institute of Higher LearningUPC Universitat Politècnica de Catalunyasaraswati vidyalaya
English, Hindi, Spanish
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (10)
cuda10
openmp10
cluster-computing10
programming-language10
c-language10
parallel-computing10
cprogramming-language10
scientific-computing10
abstraction9
cmake8
Programming languages (8)
Github contributions (5)
kokkos/kokkos
Nov 2019 - Sep 2022
Kokkos C++ Performance Portability Programming Ecosystem: The Programming Model - Parallel Execution and Memory Abstraction
Role in this project:
Back-end Developer
Back-end DeveloperContributions:268 reviews, 21 commits, 88 PRs in 2 years 10 months
Contributions summary:Rahulkumar primarily contributed to the Kokkos performance portability programming ecosystem, focusing on C++ code related to CUDA and OpenMPTarget backends. They worked on implementing and refining the parallel execution model, specifically addressing hierarchical parallelism and reductions within the OpenMPTarget implementation. The contributions also involved adapting and optimizing core functionalities, including deep copy operations and atomic load implementations, to improve performance. The user further addressed build system and testing infrastructure by updating CMake files.
cppparallel-executionprogramming-modelparallel-computingc-plus-plus
rahulgayatri23/FortKernels-
Mar 2017 - Sep 2018
Contributions:279 pushes, 10 branches in 1 year 6 months
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.