Balint Joo
Senior Developer Technology Engineer at NVIDIA
Oak Ridge, Tennessee, United States
Summary
👤
Senior🎓
Top SchoolBalint Joo is a Senior Developer Technology Engineer with 16 years of hands-on experience in high-performance computing, specializing in C/C++ MPI programming for lattice QCD and long-standing expertise with QDP++ and Chroma. He has led teams and projects at national labs and research institutions—most recently at NVIDIA and Oak Ridge—bridging physics and computer science to deliver production-ready HPC software. Balint has contributed to widely used performance-portability tooling like Kokkos by improving build and release automation across architectures, including AMD GPU support. His interests in heterogeneous computing, multi-scale algorithms, and domain-specific languages reflect a rare combination of deep numerical physics background (PhD in theoretical physics) and practical systems engineering. He is known for translating domain research into robust, portable code and for smoothing the interaction between scientific users and compiler/build ecosystems.
16 years of coding experience
20 years of employment as a software developer
PhD, Theoretical Physics, PhD, Theoretical Physics at The University of Edinburgh
Stackoverflow
11reputation
320reached
0answers
1question
Github Skills (11)
cluster-computing10
c-language10
cmake10
high-performance10
cprogramming-language10
parallel-computing10
scientific-computing10
accelerated-computing10
automation9
automations9
amdgpu8
Programming languages (9)
Github contributions (5)
kokkos/kokkos
May 2017 - Oct 2021
Kokkos C++ Performance Portability Programming Ecosystem: The Programming Model - Parallel Execution and Memory Abstraction
Role in this project:
Automation Engineer / Build & Release Engineer
Automation Engineer / Build & Release EngineerContributions:19 reviews, 7 commits, 3 PRs in 4 years 5 months
Contributions summary:Balint focused on improving the build process and adapting the project for various build environments. Their contributions included modifying CMake configuration files to ensure correct header file installation, especially for submodule builds. They also updated the build system to support different architectures and incorporated changes related to the AMD GPU architecture, as well as added the ability to install headers into a subdirectory when a submodule is used. The user further adjusted the build process to account for both standalone and submodule builds.
memorympic-plus-plusmulti-threadingkokkos
JeffersonLab/chroma
Nov 2009 - May 2022
The Chroma Software System for Lattice QCD
Contributions:610 commits, 30 PRs, 213 pushes in 12 years 7 months
qcdlatticelattice-qcdsystem-softwarechroma
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.