I am a machine learning researcher building statistical methods to enhance our grasp of biological systems in neuroscience and the life sciences more broadly. My main interest is to understand how machine learning systems can be used to infer computations and dynamics underlying a phenomenon using observational and interventional data.

In my research group at Helmholtz Munich, we develop machine learning algorithms for representation learning and inference of nonlinear system dynamics, study how large and multi-modal biological datasets can be compressed into foundation models, and study their mechanistic interpretability. If you are interested joining us, have a look at our current openings. We are currently recruiting through MCML (Research Area A3, Computational Models).

I am also very active in computing education for more than 10 years now. In 2019, I founded KI macht Schule ("AI at schools"), a non-profit organization teaching ML basics to high-school students. We provide teachers with modern teaching materials, AI tools and infrastructure in our open teaching hub, and offer courses on AI for students and teachers in Germany, Switzerland and Austria. We have a great network of volunteers in more than 9 cities who do science outreach directly in schools. We also have a growing team of instructional designers, software engineers and AI trainers to extend our platform, build new teaching materials and conduct teacher trainings. If you want to help educating the next generation of students and make them literate in AI, consider to join our team!

Finally, I am interested in deploying ML solutions in the real world. In 2023, I co-founded Kinematik AI, a company offering customized machine intelligence solutions in the biopharma and animal healthcare sector. If you have a business usecase for our customized behavioral analysis software, please reach out.

Here are some additional pointers to my work:

• 🧑‍🎓 Interested in my research? Have a look at our lab homepage or my google scholar profile.
• 🦓 Check out cebra, our new representation learning algorithm to obtain embeddings of jointly recorded behavioral & neural data.
• ☕ Check out robusta, our library for robustness & adaptation.
• 🎒 Check out how we teach ML & AI to highschool students at KI macht Schule.
• 💼 Get in touch if you are interested in what we are building at Kinematik AI.
• 🐦 Follow me on X: @stes_io
• 🐘 ... and mastodon: @[email protected]