项目介绍

2026年8月27日

In SEQUORA, we develop trustworthy AI assistance for gynecological precision oncology. Building on longitudinal clinical data, established molecular tumorboards and interoperable data models, we combine Large Language Models (LLMs), Retrieval-Augmented Generation (RAG) and knowledge-based methods into agent-based systems that support therapy-sequence decisions, clinical-trial matching and evidence synthesis in real-world tumorboard workflows. All components will be released as open-source software and reusable research artifacts.

Your tasks:

You will contribute to the clinical-informatics and AI workstream of SEQUORA:

• Co-design and implement the clinical-informatics components of the SEQUORA research prototype together with medical informaticians, gynecologic oncologists and consortium partners
• Develop and harmonize longitudinal clinical data models for more than 3,500 treatment courses, aligned with the OMOP Common Data Model and robust data-quality rules
• Develop and evaluate agent-based AI functions for therapy-sequence support, clinical-trial matching and evidence dossiers using LLM, RAG and knowledge-based methods where appropriate
• Translate clinical questions, Target Trial specifications, guidelines, literature evidence and complex trial eligibility criteria into machine-processable representations and safe clinical outputs
• Implement traceability, uncertainty communication, non-applicability rules and human-oversight mechanisms, and validate system outputs in retrospective cases and prospective tumorboard workflows
• Publish research results (minimum three peer-reviewed publications), contribute reusable open-source research artifacts and present at international conferences

Your profile:

• An excellent Master’s degree in Computer Science, Medical Informatics, Health Informatics, Data Science, or a closely related field
• A strong academic record demonstrating research excellence and the ability to pursue a doctoral degree at TUM
• Documented experience in machine learning and/or natural language processing with deep knowledge of LLMs, RAG and agent-based AI systems
• Strong Python programming skills and familiarity with machine-learning frameworks (e.g., PyTorch), data engineering, SQL and version control
• Interest in longitudinal clinical data, clinical terminologies and interoperable data models; experience with OMOP CDM or Target Trial methodology is an advantage
• A careful approach to evaluation, reproducibility, Responsible AI, uncertainty and patient safety in clinical decision-support systems
• A creative, analytical and solution-oriented mindset, with enthusiasm for interdisciplinary research in personalized medicine and women’s health
• Strong communication skills in English; proficiency in German (at least B2 level) is required for collaboration in clinical workflows

What we offer:

• The opportunity to shape clinically relevant AI research with high societal impact in gynecological precision oncology
• A fully funded PhD position (E13 TV-L, full-time, 36 months) with a structured doctoral pathway
• Joint interdisciplinary work between the Chair of Medical Informatics and the Center for Personalized Medicine, with close collaboration across the SEQUORA consortium
• Direct involvement in the design and clinical evaluation of trustworthy AI assistance in real-world tumorboard workflows
• Access to unique longitudinal clinical datasets, established molecular tumorboards and high-performance computing infrastructure (including NVIDIA B300 GPUs)
• Funding for open-access publications, international conferences and research mobility grants
• Support from the TUM Graduate School for internationalization and career development
• Employee benefits: EGYM Wellpass, corporate discounts (e.g., Käfer), and bike leasing (JobBike Bavaria)
• Free use of the library through a branch of the Munich City Library located in the building
• Working in the heart of Munich at Max-Weber-Platz with very good accessibility by public transport such as the subway, S-Bahn or tram
• Company pension plan through the Federal and State Pension Institute (VBL)
• Personal fulfillment through a varied and professionally demanding role with interdisciplinary collaboration