Hans Pabst

Zurich, Zurich, Switzerland
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

🤩
Rockstar
Hans Pabst is an application and research engineer with 12 years’ experience enabling open-source scientific applications to exploit cutting-edge HPC infrastructure and upcoming hardware in Intel’s Extreme Computing Software and Systems Group. He blends applied mathematics, algorithm design, and computational science with practical engineering from a background at Intel Labs where he worked on compiler-based AI stacks, low-level ML primitives, and hardware–software co-design. Hans contributes to prominent open-source projects—most notably improving build and packaging for libxsmm in Spack and stabilizing PlaidML’s LLVM/LIBXSMM build and CI—showing a strength in build systems, CI/CD and performance-focused integration. Based in Zurich, he pairs deep research thinking with hands-on DevOps and build engineering to make high-performance kernels and toolchains reliably deployable across diverse HPC environments. An uncommonly practical researcher, he often surfaces small but critical build and configuration changes that unlock significant performance and portability gains.
code12 years of coding experience
github-logo-circle

Github Skills (18)

package-management10
build-system10
cmake10
makefile10
cicd10
build-automation10
llvm9
c-language9
cprogramming-language9
scientific-computing8
python8
fortran8
package-manager-tool7
boost7
build-tools7

Programming languages (13)

JavaC++CTwigGoMLIRFortranDockerfile

Github contributions (5)

github-logo-circle
plaidml/plaidml

Aug 2022 - Dec 2022

PlaidML is a framework for making deep learning work everywhere.
Role in this project:
userDevOps Engineer & Automation Engineer
Contributions:1 review, 80 commits, 4 PRs in 3 months
Contributions summary:Hans primarily focused on configuring the build process and dependencies for the project. Their contributions involved updating the LLVM and LIBXSMM dependencies, which impacted the build environment. They introduced environment variables for managing LIBXSMM and made modifications to the CMake build files. Additionally, they worked on the CI/CD pipeline, specifically in the `ci/plan.py` and `ci/openvino.py` configurations, demonstrating responsibility for the build system and CI/CD integrations.
pytorchtvmdeep-learningmachine-learningcompiler
spack/spack

Mar 2017 - Dec 2021

A flexible package manager that supports multiple versions, configurations, platforms, and compilers.
Role in this project:
userBack-end Developer & Build Engineer
Contributions:1 review, 18 commits, 17 PRs in 4 years 9 months
Contributions summary:Hans primarily contributed to the `libxsmm` package within the `spack/spack` repository by adding, updating, and managing different versions of the library. They modified the package's build process, including compiler settings and variant configurations, and made changes to file installation and documentation. The user also refactored the package to use `MakefilePackage` and adjusted for file set changes in newer library versions.
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.
Request Free Trial
Hans Pabst