Akshay Venkatesh
Senior Software Engineer at NVIDIA
California, United States
Summary
🤩
Rockstar🎓
Top SchoolAkshay Venkatesh is a Senior Software Engineer based in California with 11 years of experience building high-performance, GPU-aware communication software for HPC and data-science ecosystems. At NVIDIA he leads GPU-specific additions to the UCX runtime, optimizing data paths across NVLINK, InfiniBand and shared-memory to accelerate projects like OpenMPI and RAPIDS. His background includes a PhD in Distributed Computing from Ohio State, where he worked on MPI routines for accelerators and energy-aware communication strategies. A hands-on systems developer, he implemented CUDA IPC support in the prominent open-source UCX project, contributing performance optimizations, bug fixes, and memory/event-handling refactors. Colleagues rely on him for pragmatic algorithmic choices that scale across heterogeneous clusters, and he combines deep research experience with production-grade engineering.
11 years of coding experience
5 years of employment as a software developer
Doctor of Philosophy - PhD, Distributed Computing, High Peformance Computing, Doctor of Philosophy - PhD, Distributed Computing, High Peformance Computing at The Ohio State UniversityNational Institute of Technology Karnataka
Github Skills (8)
c1710
cuda10
c-language10
c1110
cprogramming-language10
networking9
hpc9
infiniband8
Programming languages (8)
Github contributions (5)
openucx/ucx
Apr 2018 - Jan 2023
Unified Communication X (mailing list - https://elist.ornl.gov/mailman/listinfo/ucx-group)
Role in this project:
Back-end Developer
Back-end DeveloperContributions:442 reviews, 33 commits, 179 PRs in 4 years 9 months
Contributions summary:Akshay primarily contributed to the implementation of the CUDA IPC (Inter-Process Communication) functionality within the UCX (Unified Communication X) library. Their work involved adding CUDA-IPC UCT implementations, modifying and refactoring code related to memory handling, event management, and stream operations, specifically focusing on the `cuda_ipc_iface.c`, `cuda_ipc_ep.c`, and `cuda_ipc_md.c` files. The contributions also include code style improvements, bug fixes, and performance enhancements, such as caching memory handles and optimizing event handling. These changes aimed to improve the performance and efficiency of UCX's CUDA-based communication capabilities.
craympiopenshmemyangroce
Akshay-Venkatesh/ucx
Oct 2017 - Apr 2025
Unified Communication X (mailing list - https://elist.ornl.gov/mailman/listinfo/ucx-group):
Contributions:36 PRs, 1371 pushes, 210 branches in 7 years 7 months
govinfinibandmailmannetwork-toolsmailing-list
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.