Houjun Tang
Computer Research Scientist at Berkeley Lab
Berkeley, California, United States
Summary
🤩
Rockstar🎓
Top SchoolHoujun Tang is a Computer Research Scientist based in Berkeley with 12 years of experience building high-performance scientific software and infrastructure. At Berkeley Lab he has progressed from summer intern to research scientist, applying his PhD-level expertise in computer science to scalable I/O and data-management problems. He is an active open-source contributor to heavyweight projects like Spack and AMReX, adding package support, HDF5 plot/particle I/O, compression optimizations, and fixes that improve MPI and multi-rank robustness. Houjun combines research rigor with practical engineering, focusing on reproducible scientific workflows and maintainable backend systems that ease deployment of complex simulation stacks. His background spans national labs and academia, giving him a rare blend of production-grade software development and domain-aware research.
12 years of coding experience
8 years of employment as a software developer
Doctor of Philosophy (PhD), Computer Science, Doctor of Philosophy (PhD), Computer Science at North Carolina State UniversityBachelor of Engineering (B.Eng.), Computer Science, Bachelor of Engineering (B.Eng.), Computer Science at Shenzhen University
Chinese, English
Github Skills (16)
hdf10
mpi10
cluster-computing10
package-manager10
high-performance10
parallel-computing10
amr10
spack10
package-management10
python10
scientific-computing10
accelerated-computing10
cprogramming-language9
c-language9
cmake9
Programming languages (9)
Github contributions (5)
AMReX-Codes/amrex
Oct 2019 - Mar 2022
AMReX: Software Framework for Block Structured AMR
Role in this project:
Back-end Developer
Back-end DeveloperContributions:4 reviews, 19 commits, 13 PRs in 2 years 5 months
Contributions summary:Houjun primarily contributed to enhancing the AMReX software framework by implementing features related to HDF5 plot file writing. They added support to write plot files and particle data in the HDF5 format, thereby improving the framework's data output capabilities. The user also modified the existing HDF5 benchmark code to utilize the new HDF5 write implementation, and added several optimizations including ZFP and SZ compression. Furthermore, they addressed a bug in the boxcenter data and implemented a new HDF5 schema with separate datasets for different variables, as well as fixes to handle cases where MPI ranks may have no boxes.
amrmpistructuredsnl-applicationsmultiphysics
spack/spack
Jul 2021 - Dec 2022
A flexible package manager that supports multiple versions, configurations, platforms, and compilers.
Role in this project:Back-end Developer
Back-end DeveloperContributions:21 reviews, 4 commits, 10 PRs in 1 year 4 months
Contributions summary:Houjun primarily contributed to the Spack package manager by adding and updating packages for scientific software. They added new packages such as `pdc` and `sw4`, and updated existing packages like `hdf5-vol-async`. The user also addressed compilation issues and integrated new versions, demonstrating a focus on maintaining package compatibility and expanding the available software options within the Spack ecosystem. These changes involved modifications to `package.py` files, indicating work related to package definitions and dependencies.
compilerspythonradiussplatformslinux
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.