I am Jan and I love automation, because it helps me to focus on the important things in life.
Can we build a machine which autonomously discovers novel sustainable materials? This is the aim of the Materials Informatics group I lead at the Max Planck Institute for Sustainable Materials. We develop machine learning models for ab-initio thermodynamics workflows using uncertainty quantification to achieve quantitative predictions which can be directly compared with experiments.
This approach is based on our experience of developing the pyiron workflow framework, followed by the Python Workflow Definition and finally executorlib our solutions for up-scaling and rapid prototyping simulation workflows.
If you are an aspiring PhD student who wants to join us in the quest of automating materials discovery, please apply via the International Max Planck Research School for Sustainable Metallurgy (IMPRS SusMet).
Beyond my professional interest, I aim to automate repetitive tasks in my daily work, incuding:
- Two-factor authentication wth pyauthenticator
- Sorting emails into folders with gmailsorter
- Software packaging with the conda-forge community