This postdoctoral research position at ETH Zürich focuses on the development, validation, and safe integration of locally hosted Large Language Models (LLMs) for automated coding of pediatric diagnoses from electronic health records (EHRs). The project aims to enhance research capabilities and clinical data usability by leveraging advanced AI techniques in a hospital setting. The successful candidate will lead efforts in adapting and developing language models for diagnostic coding, rigorously validating models on new benchmarks, and ensuring safe deployment within clinical environments.

The position is embedded in a close collaboration between ETH Zurich’s Medical AI Lab and the University Children’s Hospital Basel (UKBB), providing direct interaction with clinicians, clinical coders, and hospital IT infrastructure. The research group is part of the Department of Biosystems Science and Engineering (D-BSSE) in Basel and is actively engaged with the ETH AI Center and SwissAI initiative, offering access to a vibrant AI community and cutting-edge computational resources, including large GPU clusters.

Applicants must have a PhD in Computer Science, Medical AI, Medical Informatics, or a related field. Essential skills include strong programming abilities in Python, experience with ML/AI/LLM stacks (PyTorch, HuggingFace, Ollama, vllm, distributed training), and proficiency in German at C1 level or higher for analyzing local patient records. Prior experience with LLMs, clinical NLP, retrieval augmented generation (RAG), vector databases, and medical data coding (ICD-10) is highly desirable. Familiarity with Docker, local server environments, GPU-based infrastructure, and good computational engineering practices is expected. Candidates should demonstrate independence, teamwork, and effective communication in English and German.

ETH Zürich is renowned for its excellence in science and technology, offering a highly interdisciplinary environment and promoting diversity, equality, and sustainability. The university provides a supportive atmosphere for research, teaching, and professional growth, with a commitment to climate-neutral operations and equal opportunities for all staff and students.

To apply, candidates must submit their application online via the ETH Zurich portal, including a single PDF with CV, Bachelor and Master transcripts, motivation letter, and letters of recommendation (or a list of referees). Applications via email or postal services will not be considered. For further information or questions about the position, contact Prof. Michael Moor's lab email (no applications).