Senior AI Engineer - Imaging

About the role

Roche is looking for a senior imaging-focused machine learning engineer within its AI Enablement team in Basel. You will work alongside computational scientists, biologists, bench scientists, AI/ML experts, product leaders and software engineers, building scalable AI/ML and computer vision systems that turn scientific imaging data (microscopy, digital pathology, biomedical imaging) into quantitative analysis and richer phenotypic insights.

Key responsibilities

• Design, develop and test robust, scalable AI/ML scientific web applications and backend pipelines for high-throughput imaging data processing and visualization.
• Build, evaluate and fine-tune computer vision and deep learning models such as cell segmentation, cell classification, phenotypic profiling, tissue segmentation, feature extraction and image registration.
• Develop tools to evaluate model performance, set up benchmark datasets and track quality, robustness and drift over time.
• Maintain cloud infrastructure and data pipelines for large-scale imaging datasets (e.g. OME-TIFF, Zarr, HDF5).
• Contribute to architecture, code reviews and the evolution of engineering and MLOps processes.

Required profile

• PhD, Master's with 3-5 years, or Bachelor's with 4-7 years of experience in a technical field with a focus on machine learning engineering and computer vision.
• Demonstrated experience with high-dimensional scientific imaging data (fluorescence/brightfield microscopy, whole slide images, high-content screening, multiplexed imaging).
• Experience with multi-modal data fusion or in a pharma/biotech research environment.
• Excellent communication skills and ability to partner with scientific collaborators and cross-functional teams.

Required skills

• Strong proficiency with AI/ML and computer vision libraries such as PyTorch, TensorFlow, OpenCV, MONAI, Scikit-Image, torchvision and Napari.
• Deep expertise in CNNs, Vision Transformers, UNet, SAM and Cellpose.
• Knowledge of Kubernetes, relational and NoSQL databases, object storage and cloud-native architectures (ideally AWS).
• Nice to have: workflow automation, GenAI, Vision-Language Models and AI agents.