James Elliott
Principal Member Of Technical Staff at Sandia National Laboratories
Albuquerque, New Mexico, United States
Summary
🤩
Rockstar🎓
Top SchoolJames Elliott is a Principal Member of Technical Staff at Sandia National Laboratories with nine years of experience bridging high-performance computing research and production-grade performance engineering. He leads application performance efforts for heterogeneous and novel architectures, represents Sandia across DOE centers of excellence, and has a strong research pedigree from a PhD focused on resilient numerical methods and "Skeptical Programming" for preconditioned iterative solvers. His open-source contributions include performance-focused backend work on the widely used Kokkos C++ performance portability ecosystem, where he optimized deep copy paths for modern architectures using OpenMP and memcpy-based chunking. James combines rigorous academic training in computer science and computational analysis with hands-on systems tuning, and is known for improving both numerical robustness and build/runtime efficiency in complex codebases.
9 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy (Ph.D.), Computer Science, Doctor of Philosophy (Ph.D.), Computer Science at North Carolina State UniversityMaster of Science (M.S.), Mathematics and Statistics, Master of Science (M.S.), Mathematics and Statistics at Louisiana Tech University
English
Github Skills (9)
openmp10
cluster-computing10
cpp10
memcpy10
parallel-computing10
cplus10
scientific-computing10
k9
programming-language8
Programming languages (5)
Github contributions (5)
kokkos/kokkos
Jan 2019 - Feb 2019
Kokkos C++ Performance Portability Programming Ecosystem: The Programming Model - Parallel Execution and Memory Abstraction
Role in this project:
Back-end Developer / Performance Engineer
Back-end Developer / Performance EngineerContributions:1 review, 8 commits, 1 PR in 19 days
Contributions summary:James focused on optimizing deep copy operations within the Kokkos library, a performance portability framework. Their contributions involved modifying and parallelizing the deep copy implementation using OpenMP and `memcpy`. They addressed issues related to page alignment and chunking to improve performance on different architectures like KNL and HSW, while also incorporating code improvements suggested by other developers. Additionally, they refactored the code by moving the implementation into a separate cpp file for faster build times.
memorympic-plus-plusmulti-threadingkokkos
jjellio/spack
Mar 2021 - Oct 2021
A flexible package manager that supports multiple versions, configurations, platforms, and compilers.
Contributions:63 pushes, 12 branches in 7 months
compilersplatformspackage-managerdarwinconfigurations
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.