I am a relAI PhD student at Helmholtz Munich and the Technical University of Munich (TUM), supervised by Dr. Vincent Fortuin. I aim to contribute to the reliable application of AI in science through better uncertainty quantification and robustness.

Previously, I worked as research associate in applied AI for the energy transition. I focused on reliable wind power forecasting and developed software for more efficient and streamlined ML development in the energy research community. Before that, I studied mathematics at TUM, focusing on probability theory, statistics, and financial applications.

I strongly believe Bayesian deep learning is a key approach to reliable AI for science. My PhD project aims to investigate how Bayesian principles can be applied safely in modern ML, how models can be informed with prior knowledge, and how this leads to more reliable and data-efficient ML for scientific tasks.