Jacob Faibussowitsch
Senior Software Engineer, Distributed Task-based Runtimes at NVIDIA
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolJacob Faibussowitsch is a Senior Software Engineer specializing in distributed, task-based runtimes and high-performance scientific computing, currently at NVIDIA with nine years of experience across national labs and academia. He combines deep C/C++ and GPU expertise (CUDA/HIP, MPI) with practical performance engineering on flagship libraries like PETSc and MFEM, including contributions that unify GPU backends and optimize core numerical kernels. While pursuing a Ph.D. in Mechanical Engineering at UIUC, he has driven GPU-aware MPI integration and co-authored work on asynchronous GPU streams for PETSc presented at SIAM CSE. His background spans end-to-end research software development—from porting linear algebra and multigrid smoothers to device architectures to coupling tight-binding DFT code with LAMMPS for MD workflows. Jacob is notable for bridging research and production: he authors portable device context abstractions and performance-driven low-level changes that scale on exascale-class systems. Outside core contributions, he maintains an active open-source presence, publishing toolkits and performance patches that reveal a preference for pragmatic, reproducible engineering.
9 years of coding experience
1 year of employment as a software developer
University of Illinois Urbana-Champaign
English, German, Spanish
Github Skills (18)
c-language10
cluster-computing10
c1110
c1710
parallel-computing10
performance-optimization10
cuda10
finite-element-analysis10
cprogramming-language10
scientific-computing10
computational-science9
data-structure9
hpc9
numerical-methods9
data-structures9
Programming languages (14)
Github contributions (5)
petsc/petsc
Sep 2019 - Jan 2023
Mirror of https://gitlab.com/petsc/petsc
Role in this project:
Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:354 commits in 3 years 5 months
Contributions summary:Jacob contributed to the core functionality of the PETSc library, specifically focusing on performance improvements and low-level code optimization. Their work involved fixing spelling errors, enhancing printing functionalities, and adding capabilities for handling different data types. The user also engaged in in-depth modifications of core routines within the library, demonstrating an understanding of numerical methods and data structures.
petsc
mfem/mfem
Jun 2021 - Sep 2021
Lightweight, general, scalable C++ library for finite element methods
Role in this project:Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:6 reviews, 37 commits, 3 PRs in 3 months
Contributions summary:Jacob primarily contributed to performance optimizations and device porting within the `mfem` repository, a finite element methods library. They modified code to integrate with the `cusparse` library, specifically checking for the version of `cusparse` to select the appropriate algorithm. Additionally, they ported `jacobi2` and `jacobi3` to the device and addressed style issues within the code. Their work focused on leveraging GPU capabilities and ensuring code correctness, enhancing the library's performance.
cppfinite-elementshigh-orderhigh-performance-computingparallel-computing
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.