PhD Position in Foundation Models Medical Reasoning

PhD Position in Foundation Models for Medical Reasoning

The research group led by Michael Moor is looking to fill a PhD position focusing on medical reasoning with foundation models. We are looking for an ambitious candidate eager to conduct groundbreaking research in medical AI. Our goal is to advance the capabilities and faithfulness of medical reasoning models across diverse medical contexts. The successful candidate will join a cutting-edge research group that is pioneering medical foundation models, agent systems, and reasoning models.

Develop novel methods that advance the reasoning capabilities of medical foundation models (LLMs, MLLMs, and multi-agent systems).

Design multimodal retrieval-augmentation strategies integrating EHRs, clinical guidelines, imaging databases, and other structured knowledge sources.

Implement flexible memory and retrieval systems to support reasoning-chain generation and sequential decision-making in clinical tasks.

Develop approaches to reduce hallucinations and improve reliability and robustness of medical AI systems, including adversarial robustness (in collaboration with SHS).

Post-train and test-time scale multimodal foundation models at multiple scales.

Co-develop a new benchmark for multimodal clinical reasoning and explainable sequential decision-making.

Publish research results in top-tier ML venues (e.g., NeurIPS, ICML, ICLR) and top-tier biomedical journals (e.g., Nature Medicine, Nature Biomedical Engineering).

Participate in collaborations across the MLCARE consortium (including funded secondments to a Max Planck Institute and Siemens).

Engage in dissemination, open-source contributions, and knowledge-transfer activities.

Master’s degree in Computer Science, AI, Machine Learning, Mathematics, Electrical Engineering
, or or

Master’s degree in Medicine (MD) with strong Python skills and some ML experience.

Strong programming skills in Python and experience with modern ML stacks (PyTorch, Hugging Face, distributed training).

Experience in LLMs, vision–language models, multimodal learning, or clinical NLP is highly welcome.

For computational candidates: experience in large-scale ML training (multi-node setups, distributed training) is a strong advantage.

Good computational engineering practices: version control, reproducible pipelines, batch job management, etc.

Ability to work independently, contribute to team efforts, and communicate effectively in English.

Prior research experience, publications, or industry ML experience are pluses but not required
.

• A full-time, fully funded MSCA Doctoral Network position at ETH Zurich
  , one of the world’s leading research universities.
• Our group is engaged with the ETH AI Center and Swiss
  
  AI initiative, giving our group members access to this vibrant and world-class AI community.
• Access to cutting-edge computational resources, including large GPU clusters, and medical and biomedical collaborations.
• Access to the Europe-wide MLCARE consortium for exchange and collaborations.
• Employment in a highly interdisciplinary environment at Department of Biosystems Science and Engineering (D-BSSE) located in Basel, embedded in a major hub for medical and biomedical research and biotechnology.

In line with our values , ETH Zurich encourages an inclusive culture. We promote equality of opportunity, value diversity and nurture a working and learning environment in which the rights and dignity of all our staff and students are respected. Visit our Equal Opportunities and Diversity website to find out how we ensure a fair and open environment that allows everyone to grow and flourish.

Sustainability is a core value for us – we are consistently working towards a climate-neutral future .

We look forward to receiving your online applicationincluding the following documents (concatenated into one PDF):

• CV
• Bachelor and Master transcripts
• Motivation letter (motivation & fit to the program and the host lab)
• Letters of recommendation (if available, also just a list of names that can be queried for letters of recommendation would suffice)

Please note that we exclusively accept applications submitted through our online application…