PhD Position in Trustworthy Biomedical AI (f/m/x)

102986

Full time
39 hrs./week

Neuherberg near Munich

Partial Home Office possible

At Helmholtz Munich, we develop groundbreaking solutions for a healthier society in a rapidly changing world. We believe that diverse perspectives drive innovation. Through strong partnerships, we accelerate the transfer of new ideas from the lab to real-life applications, improving lives.

, the artificial intelligence platform of the Helmholtz Association, aims to democratize AI and enhance its capabilities to address cutting-edge scientific questions and major societal challenges. As an application-driven platform, we accelerate scientific discovery across the Helmholtz Association by enabling the development and implementation of AI solutions.

We host dedicated research groups and consultancy teams that are at the heart of our activities, ensuring the broad dissemination of AI methods and the integration of data-driven research across disciplines. We foster collaboration, ensure access to key resources and expertise, and empower researchers to harness the full potential of artificial intelligence.

The builds open, trustworthy AI for biomedicine. Our work spans three connected areas:

• Representing biomedical knowledge in forms that machines can reason over.
• Building agentic and co-scientist systems that use that knowledge.
• Developing the evaluation methods that tell us whether any of it can be trusted.

We sit at the meeting point of machine learning methods and biomedical application, embedded in Helmholtz AI, the Computational Health Center, and the German Center for Diabetes Research (DZD). We are working in the open by sharing code and methods with the research community. Our tools (BioCypher, BioChatter, BioContextAI) are used by research groups worldwide, and we hold ourselves to reproducible, well-tested, community-facing engineering.

As a PhD Candidate (f/m/x) you will define and drive your own doctoral project within the group's scope. You set the research questions and build the methods and systems to answer them. You will work in a team with individual supervision.

Your tasks

• Formulate a doctoral project around one or more of the research directions below, or a well-argued proposal of your own that fits the group's scope.
• Design, implement, and evaluate methods and systems, working openly through open-source software and reproducible research practices.
• Publish at leading ML and bioinformatics venues and present your work at international meetings.
• Contribute to the group's open-source ecosystem and the wider community around it.
• Collaborate across the method and domain expertise available in our institutional networks.

Your profile

• Education: Master's degree (obtained or near completion) in Computer Science, Machine Learning, Data Science, Bioinformatics, Computational Biology, or a related field.
• Technical skills: Strong Python and a working git, testing, and documentation habit; depending on your chosen direction, experience with machine learning, NLP, or LLMs is an advantage.
• Self-direction: Evidence that you can scope and drive work on your own (for instance, a portfolio, open-source contributions, or a thesis or side project you led yourself).
• Domain interest: Curiosity about biomedicine; a domain background is welcome but not required, we will help you build it; candidates coming from biology or medicine with solid Python skills are explicitly welcome.
• Communication: Fluent written and spoken English.
• Disposition: Comfort with open-ended problems and the ability to turn a vague question into a concrete plan.

Research directions

• Trustworthy AI and evaluation, reliability and robustness (e.g., how do we know when an agent's literature synthesis is wrong?)
• Agentic and co-scientist systems, orchestration and epistemic processes (e.g., how should a co-scientist decide what to verify before it acts?)
• Knowledge representation, grounding, representation learning (e.g., what graph or embedding structure lets an LLM ground its claims?)

This position provides the opportunity for you to build specialist knowledge whilst developing key professional experience, both of which will help you to make strides in your scientific career.

Remuneration and social security benefits are based on the German Collective Wage Agreement for Public-Sector Employees at Federal Level (EG 13 50% TV EntgO Bund). In addition, there is also the possibility of receiving an allowance if the applicable conditions are met. The position has an (initial) fixed term of three years but may be extended under certain circumstances.

We offer:

1. A group where you own your research and work in the open.
2. Integration into Helmholtz AI, the Computational Health Center, and the DZD network, with access to compute, curated biomedical corpora, and knowledge graphs.
3. The chance to contribute to open-source frameworks and real deployments used well beyond our lab.
4. Close mentoring from the group and structured graduate-school training toward a TUM doctorate.
5. Salary according to TVöD E13 (75%) with social benefits and flexible remote work arrangements.

Benefits

• Career Development: Postdoc program, scientific training & career center with tailored offers
• Scientific Resources: State-of-the-art infrastruture and Core Facilities
• Recreation: 30 days annual leave, flexi days, plus public holidays
• International Staff Service: Support with the relocation and integration process in Germany
• Health Promotion: Sports, company doctor, mental health initiatives

Interested in applying?

If you have any questions, feel free to contact Sebastian Lobentanzer, sebastian.lobentanzer@helmholtz-munich.de, who will be happy to help.

Our recruiting is decentralized - your application will be reviewed directly by the specialist department in which you could work in the future.

Please send your application only via our online application tool and with the following documents:

• CV
• Cover letter
• Degrees/Diplomas/Certificates
• Contact details for at least two referees
• A short motivation letter (1 page max.) that includes a brief sketch of a project you would want to pursue in the group, naming a concrete biomedical problem or use case (a rough idea is what we are looking for, not a polished proposal; this sketch matters more to us than a perfect record; you may use generative AI, but please make sure you take full responsibility for the proposed plan)
• A transcript of all degrees obtained
• Links to any software repositories or portfolio materials

If you have obtained a university degree abroad, we will require further documents from you regarding the comparability of your degree by the time you start work at the latest.

We are committed to promoting a culture of diversity and welcome applications from talented people regardless of gender, cultural background, nationality, ethnicity, sexual identity, physical abilities, religion or age. Qualified applicants with physical disabilities will be given preference.

Our commitment

Helmholtz Munich
Deutsches Forschungszentrum für Gesundheit und Umwelt (GmbH)
Helmholtz AI
Ingolstädter Landstraße 1
85764 Neuherberg