Design and implement downstream AI models that operationalize clinical endpoints using outputs from foundation models.
Translate clinical study designs and outcome definitions into clear modeling tasks and evaluation frameworks in collaboration with Clinical Scientists.
Define and apply consistent modeling and evaluation approaches across multiple biomarkers and imaging modalities.
Collaborate closely with the AI Algorithm Lead to ensure robust integration between foundation models, MLOps infrastructure, and downstream biomarker models.
Guide and review modeling approaches developed by AI Research Scientists, providing technical feedback and mentorship.
Perform in-depth model analysis, including calibration, subgroup performance, and failure-mode assessment.

MSc or PhD in Computer Science, Machine Learning, Biomedical Engineering, Applied Mathematics, Physics, or a related technical field
At least 5 years of experience developing AI or machine learning models in a medical imaging or clinical research context
Proven ability to translate clinical or scientific questions into appropriate modeling approaches (e.g. classification, risk prediction, longitudinal modeling)
Experience working with multimodal data (e.g. imaging combined with clinical or pathology data)
Strong understanding of model evaluation, calibration, robustness, and subgroup performance in real-world datasets
Familiarity with foundation models and downstream fine-tuning or adaptation strategies
Proficiency in Python and deep learning frameworks, and experience working in Linux-based environments
Preferred qualifications: Experience in NLP with LLMs or VLMs

Certificate of Conduct (VOG) or background check is part of application procedure

AI Research Engineer

Key skills