Postdoc in Fair AI & Oculomics for Global Health (f/m/d)

Tasks

• Develop and evaluate algorithmic strategies to detect and mitigate bias in foundation models using African fundus datasets
• Design and implement interpretability techniques applicable to foundation models
• Co-develop deep learning pipelines to reconstruct systemic health conditions and histories from fundus images
• Oversee high-quality data collection and data exchange between The Gambia and University of Tübingen
• Co-lead the project's strategic direction, co-supervise a PhD student, and disseminate findings

Profile

• Doctoral degree in computer science, medical informatics, physics, mathematics, or a related quantitative field with a focus on machine learning
• Proven track record in developing and implementing deep learning algorithms, specifically for medical image analysis (experience with foundation models or interpretability is a strong plus)
• Advanced programming skills in Python and experience with deep learning frameworks such as PyTorch, including handling large-scale datasets
• Experience (or strong potential) in co-leading international research projects and the ability to coordinate complex data-sharing workflows across borders
• Strong interest in Global Health and the willingness to travel to Gambia for data collection coordination; excellent communication skills in English are essential

About the Faculty of Medicine

The Faculty of Medicine is one of the four founding faculties of the Eberhard Karls University of Tübingen. With its non-clinical facilities as well as its research and teaching area corresponding to the organisational units of the University Hospital, it is one of the largest medical training and research institutions in Baden-Württemberg.

About Hertie AI

The “Hertie Institute for AI in Brain Health” (Hertie AI) at the Faculty of Medicine of the Eberhard Karls University Tübingen is funded by the Gemeinnützige Hertie Stiftung to develop AI algorithms that detect diseases of the nervous system earlier and treat them better. The postdoctoral position is offered within the framework of the DFG-funded project “De-biased and interpretable foundation models to reconstruct health history from fundus images in The Gambia “ between Hertie AI and the MRC Unit in The Gambia.

Contact for questions

Mr. Prof. Philipp Berens
Hertieai@medizin.uni-tuebingen.de

Online application to

Mr. Prof. Philipp Berens
Index number: 7258
Including CV and cover letter


Application deadline: 08.03.2026