100% / Available: February 2026

Neural network models have transformed many areas of life sciences, including protein structure prediction and molecular generation. However, due to limited high-quality data, purely data-driven AI models often lack the generalizability required to reliably model protein–ligand interactions, as recently demonstrated by our group (https://doi.org/10.1038/s41467-025-63947-5).
Our research therefore focuses on advancing next-generation drug design methodologies by integrating physicochemical principles directly into deep neural network approaches. Representative publications from our group include:

Your position

• Designing and implementing innovative deep neural network models.
• Integrating physical principles and molecular modeling knowledge into learning architectures.
• Collaborating with experimental research groups, enabling real-world validation and application of newly developed algorithms.

Your profile

• MSc in the fields of Physics, Computational Chemistry or Computer Sciences.
• Excellent knowledge in Statistical Mechanics & Thermodynamics.
• Research experience preferably with publication.
• Strong programming skills in Python.
• Experience in machine learning, in particular neural network concepts.
• Fluent verbal and written communication skills in English.
• Highly motivated, interactive team player.

We offer you

• PhD student position.
• Training into the key methods of an emerging research field.
• International and collaborative research environment.

• Letter (max. 1 page) highlighting motivation, experience and skills
• CV
• Diploma of Bachelor's and Master's degree
• Contact details of at least two academic references