Logo HZDR

Student Assistant / Research Assistant

Automated workflows for reproducible benchmarks (Id 328)

No current offer!

To establish reproducible benchmarks and data science workflows, we need support from curious collaborators with experience in software development using Python. We would like to extend the open-source Snakemake workflow engine with support for spack/pip environments. Following this task, benchmarks for inference measurement of machine learning systems on local HPC systems will be performed using the above extensions.

Department: Computational Science

Contact: Dr. Steinbach, Peter

Requirements

  • Programming with Python
  • Code organisation of a Python project (installation, code formatting, tests, integration)
  • Contribution to open source projects on Github

Conditions

  • Supervision by a team member of the Helmholtz AI Consultants
  • Remote-only work
  • Access to HPC cluster if necessary
  • follow-up projects with machine learning focus possible