Skip to content
View jfdev001's full-sized avatar
9 followers · 27 following

Achievements

Achievement: YOLOAchievement: QuickdrawAchievement: Pull Sharkx2Achievement: Pair Extraordinaire

Achievements

Achievement: YOLOAchievement: QuickdrawAchievement: Pull Sharkx2Achievement: Pair Extraordinaire

Block or report jfdev001

Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
jfdev001/README.md

Jared Frazier

Software engineer with experience in HPC environments, working on large-scale simulation software such as the ICON climate/weather model. Experienced in performance optimization, automation, and DevOps

Current interests

  • High performance computing + DevOps in these environments
  • Machine learning + climate modelling
  • Finite element/volume methods
  • Computational fluid dynamics
  • Compiler development

Free and Open Source Software (FOSS) Contributions

FTorch | Major contributions to a library for calling PyTorch models from Fortran

  • Added automatic generation of pkg-config files to simplify integration into legacy build systems (PR).
  • Enabled building FTorch as a static library, supporting integration with operational NWP models like the German Weather Service's (PR).
  • Implemented GitHub Actions CI workflows for Intel compilers, improving cross-platform build reliability (PRs).
  • Patched subtle compilation errors and improved test-suite stability (PRs)
  • Ongoing maintenance and community support (PRs, issues).

Other Open Source Contributions

  • Ferrite.jl: Improved documentation and fixed deadlinks (PRs, issues).
  • PETSc: Fixed function signature documentation and tutorial code (PRs).
  • fftpack: Sanitized CMake files (PR).
  • pFUnit: Reported edge-case failure when using pFUnit as a dependency (issues).
  • SpeedyWeather.jl: Identified reproducibility issues with library-documented examples (issues).

Pinned Loading

  1. algebraic-multigrid algebraic-multigrid Public

    Cross platform algebraic multigrid implementation using C++, Eigen3, and Catch2. For autodidactic purposes.

    C++ 3 1

  2. hello-sparse-blas hello-sparse-blas Public

    Boilerplate Modern CMake project with Fortran and Intel MKL SpBLAS. Optional unit testing and arbitrary dependency fftpack included for illustrative purposes.

    CMake 3

  3. icon-model-machine-learning-backbone icon-model-machine-learning-backbone Public

    Configuration boilerplate for using PyTorch ML models in ICON.

    Fortran 1

  4. PlasmaCellDiff PlasmaCellDiff Public

    Julia implementation of "Quantitative modeling of the terminal differentiation of B cells and mechanisms of lymphomagenesis" by Martinez 2012 (https://doi.org/10.1073/pnas.1113019109)

    Julia

  5. parallel-nbody parallel-nbody Public

    C MPI + OpenMP implementation of n-body simulation for the Parallel Programming Practical (PPP) course at Vrije Universiteit.

    C

  6. mlp-from-scratch mlp-from-scratch Public

    Implementation and extensive report on the multilayer perceptron to elucidate fundamental concepts in deep learning.

    Python