Jeff Hammond
Distinguished Engineer at NVIDIA
Helsinki, Mainland Finland
Summary
🤩
Rockstar🎓
Top SchoolJeff Hammond is a Distinguished Engineer with 18 years in high-performance computing, now building GPU communication software at NVIDIA from Helsinki. He brings deep expertise in parallel programming models, performance engineering, and computational chemistry—grounded in a PhD from the University of Chicago and long-term work on NWChem and MPI. His career spans research and industry roles at Argonne, Intel, and NVIDIA, influencing exascale co-design, MPI/PGAS standards, and GPU-centric runtimes like NCCL and NVSHMEM. An active open-source contributor, he has optimized critical projects such as Kokkos, MPICH, Spack, and BLIS, often focusing on atomic performance, portability, and test robustness. Known for practical hardware-software co-design, he helps architects turn workload analysis into usable features rather than abstract specs. He is unusually settled—deeply committed to his team and location—so he focuses on long-term impact over career moves.
19 years of coding experience
21 years of employment as a software developer
BS & BA, chemistry & mathematics, BS & BA, chemistry & mathematics at University of WashingtonMS, PhD, Chemistry, MS, PhD, Chemistry at University of Chicago
Stackoverflow
5,484reputation
458kreached
109answers
5questions
fortran
top-5%
openmp
top-5%
parallel-processing
top-5%
intel
top-5%
dynamic-memory-allocation
top-5%
mpi
top-1%
Github Skills (39)
optimization10
programming-language10
c-language10
package-management10
atomics10
quality-assurance10
mpi10
cluster-computing10
hpc10
c1110
blas10
makefile10
c1710
message-passing10
parallel-computing10
Programming languages (25)
Github contributions (5)
flame/blis
Nov 2015 - Jan 2022
BLAS-like Library Instantiation Software Framework
Role in this project:
Backend & Performance Engineer
Backend & Performance EngineerContributions:3 reviews, 15 commits, 23 PRs in 6 years 3 months
Contributions summary:Jeff primarily focused on optimizing the BLIS library by modifying build configurations and utilizing hardware-specific features. Their work involved improving performance by integrating GCC intrinsics and leveraging Intel's optimized math libraries. The user also addressed build issues and enhanced code portability across different architectures, including allowing builds for KNL without specific memory allocation libraries. These changes demonstrate a strong understanding of build systems and hardware optimization techniques.
blasinstantiationblisblas-librarieslinear-algebra
spack/spack
Dec 2017 - Jun 2025
A flexible package manager that supports multiple versions, configurations, platforms, and compilers.
Role in this project:Backend Developer
Backend DeveloperContributions:12 reviews, 8 PRs, 31 comments in 7 years 7 months
Contributions summary:Jeff primarily contributed to the Spack package manager by adding support for and making modifications to existing packages, specifically those related to high-performance computing (HPC) and scientific computing. They added new packages like ARMCI-MPI, modified the build configurations of existing packages such as BLIS and NWChem, and introduced new variants. The user appears to be focused on improving compatibility and functionality of the Spack package manager with various scientific software and hardware configurations.
compilerpackage-managerpythonspackhpc
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.