Brian Friesen
System Management & Software Group Lead at Berkeley Lab
Berkeley, California, United States
Summary
👤
Senior🎓
Top SchoolBrian Friesen is a Berkeley-based software and systems leader with a PhD in physics and 11 years of experience building and optimizing HPC systems and performance at Berkeley Lab and NERSC. He currently leads the System Management & Software Group and serves as NERSC-10 Deputy Project Director for Systems, bridging hands-on engineering, performance consulting, and large-scale system design. His background as an application performance specialist and AMReX contributor reflects deep expertise in tuning scientific frameworks for manycore architectures and improving parallel boundary-condition updates. Comfortable moving between low-level build/config tweaks and strategic program leadership, he excels at turning research-grade codes and infrastructure into production-ready, scalable solutions. Colleagues know him for combining rigorous academic training with practical, results-driven engineering across compute, monitoring, and system management domains.
11 years of coding experience
5 years of employment as a software developer
Bachelor's Degree, Physics, with honors, summa cum laude, Bachelor's Degree, Physics, with honors, summa cum laude at Oklahoma Baptist UniversityDoctor of Philosophy (PhD), Physics, Doctor of Philosophy (PhD), Physics at University of Oklahoma
English, German
Github Skills (10)
openmp10
cluster-computing10
parallel-computing10
multithreading10
scientific-computing10
computational-physics9
fortran9
mpi9
c-language8
cprogramming-language8
Programming languages (8)
Github contributions (5)
AMReX-Codes/amrex
May 2015 - Sep 2020
AMReX: Software Framework for Block Structured AMR
Role in this project:
Back-end Developer
Back-end DeveloperContributions:203 commits, 35 PRs, 2 pushes in 5 years 4 months
Contributions summary:Brian primarily contributed to the AMReX framework by adding support for the Babbage testbed at NERSC. They modified build configurations to include the correct compiler flags for the Intel MIC architecture and added run scripts for the tiled heat solver. Furthermore, the user refactored core code for performance improvements, changing the BoxArray and Geometry classes and adding analysis capabilities to existing code to exploit multithreading. These contributions improved the performance of updating periodic boundary conditions.
amrmpistructuredsnl-applicationsmultiphysics
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.