Alle Stellen
We are seeking an exceptional and highly motivated Postdoctoral Researcher to lead research on multimodal reasoning models for oncology. The project focuses on developing, post-training, and evaluating flexible AI models that can support complex oncologic diagnostic and therapeutic decision-making in a safe, transparent, and clinically grounded manner. The position is embedded within a highly interdisciplinary collaboration between ETH Zurich, Kaiko.ai, and clinical partners, offering an opportunity to advance foundational AI research while working toward real-world translation in oncology.
Tasks
- Development and adaptation of oncology-focused foundation models capable of reasoning over complex clinical questions, including diagnosis, molecular interpretation, treatment selection, and longitudinal care.
- Design and implementation of multimodal language model architectures integrating clinical context, biomedical literature, guidelines, and patient-level multimodal evidence.
- Adaptation and evaluation of models on public and institutional oncology datasets.
- Development of uncertainty-aware and safety-aware reasoning behavior.
- Creation of model workflows that use external tools and knowledge sources in a reliable and auditable way, such as retrieval from literature, clinical guidelines, and trial databases.
- Clinical trial matching and therapy evidence lookup.
- Variant interpretation and molecular knowledgebase use.
- Development of multi-agent systems for decomposing complex oncology tasks into hierarchical context streams.
- Generation of citation-grounded and traceable outputs suitable for expert review.
- Development of post-training methods to improve clinical reasoning quality, reliability, and safety, including process-level supervision, outcome-based supervision, and reinforcement learning for oncology-specific reasoning behavior.
- Comparison and development of RL training approaches.
- Calibration, abstention, and safety-aware optimization.
- Evaluation of oncology reasoning models in clinically meaningful settings, focusing on guideline concordance, diagnostic and therapeutic reasoning quality, molecular interpretation accuracy, tool-use reliability, citation quality, evidence grounding, calibration, uncertainty, appropriate deferral, trace auditability, and clinician-in-the-loop evaluation.
Requirements
- PhD in Computer Science, Machine Learning, Medical AI, Biomedical Informatics, Computational Biology, or a related field.
- Strong programming skills in Python and modern ML frameworks.
- Experience with deep learning and large language models.
- Strong publication record in AI/ML, medical AI, computational biology, biomedical informatics, or related areas.
- Ability to work in highly interdisciplinary research environments.
- Experience with foundation models, multimodal models, or biomedical/clinical language models is preferred.
- Experience with reasoning models, agents, tool use, or compound LLM systems is preferred.
- Experience with LLM post-training methods such as RLHF, RLAIF, verifier-guided training, or process supervision is preferred.
- Familiarity with retrieval methods for LLMs, including dense/sparse retrieval, agentic retrieval, or hybrid approaches is preferred.
- Experience with medical AI applications, particularly oncology, genomics, imaging, or clinical NLP is a plus, but not required.
- Experience with scalable ML infrastructure, multi-node GPU training, or local/private deployment settings is preferred.
Benefits
- A full-time postdoctoral position at ETH Zurich.
- Collaboration between ETH Zurich (D-BSSE located in Basel) and Kaiko.ai.
- Opportunity to work on cutting-edge foundation models for real-world oncology reasoning.
- Access to unique multimodal clinical datasets and close collaboration with Kaiko.ai and clinical partners.
- Highly interdisciplinary environment spanning AI (foundation models, MLLMs, agent systems), oncology, and clinical informatics.
- Competitive salary and excellent research infrastructure, including access to the Alps cluster with 10,000 high-end GPUs within SwissAI projects.
- Engagement with the ETH AI Center and SwissAI initiative, providing access to a vibrant and world-class AI community.
- ETH Zurich encourages an inclusive culture, promotes equality of opportunity, values diversity, and nurtures a working and learning environment in which the rights and dignity of all staff and students are respected.
- Sustainability is a core value, with consistent efforts towards a climate-neutral future.
Bist du Teil dieses Unternehmens?
Dieses Unternehmensprofil wurde automatisch erstellt. Wenn du für ETH Zürich arbeitest, kannst du das Profil jetzt beanspruchen und verifizieren – kostenlos und in wenigen Minuten.
Verifizierte Profile erhalten ein Siegel und können ihre Seite, Stellen und Bewerbungen direkt verwalten.
Über uns
Die ETH Zürich ist eine technisch-naturwissenschaftliche Hochschule des Bundes mit Sitz in Zürich. Wir betreiben Lehre, Forschung und Wissenstransfer in Bereichen wie Ingenieurwissenschaften, Naturwissenschaften, Architektur, Mathematik, Informatik, Management sowie Sozial- und Geisteswissenschaften. Unsere Ausbildung umfasst Bachelor-, Master-, Doktorats- und Weiterbildungsangebote und ist eng mit internationaler Spitzenforschung verbunden. Als Teil des ETH-Bereichs entwickeln wir wissenschaftliche Grundlagen, Technologien und Lösungen für Gesellschaft, Wirtschaft und globale Herausforderungen.
Das Team
The position is part of a highly interdisciplinary collaboration between ETH Zurich, Kaiko.ai, and clinical partners, and is based at the Department of Biosystems Science and Engineering (D-BSSE) in Basel. The group is actively engaged with the ETH AI Center and SwissAI initiative.