Senior Principal AI/ML Scientist, Computational Imaging

Role Summary

Senior Principal AI/ML Scientist, Computational Imaging leads development and application of advanced AI/ML methods for medical imaging, focusing on digital pathology and/or radiology. Interfaces with translational pathology, clinical imaging, clinical biomarkers, and computational biologists; mentors junior scientists and influences external collaborations. Reports to the Executive Director of Computational Imaging and Digital Biomarkers.

Responsibilities

• Lead the design, development, training, and validation of AI/ML models for digital pathology and/or radiology applications.
• Define technical direction for custom AI/ML model development, including architecture selection, training paradigms, validation strategies, and performance benchmarks.
• Develop custom deep learning architectures and workflows for segmentation, classification, representation learning, and prediction tasks.
• Leverage and adapt foundation models (e.g., vision transformers, multimodal and self-supervised models), including fine-tuning and domain adaptation using proprietary datasets.
• Extract insights from large-scale imaging datasets, including whole-slide images and radiology modalities (CT, MRI, PET).
• Apply advanced computer vision and ML methods, including CNNs, U-Net variants, Vision Transformers, diffusion-based or representation-learning models.
• Define evaluation strategies and ensure analytical rigor, reproducibility, and scientific credibility.
• Integrate imaging data with clinical, molecular, or spatial-omics data where relevant.
• Balance innovation with practicality, ensuring solutions are scalable, interpretable, and fit-for-purpose.
• Work closely with pathologists, radiologists, clinicians, and translational scientists to translate scientific questions into computational imaging solutions.
• Communicate modeling approaches, assumptions, results, and limitations to technical and non-technical stakeholders.
• Contribute to shaping project-level research questions and study designs involving imaging data.
• Support external collaborations through technical input and scientific exchange as needed.
• Contribute to the organization’s scientific visibility through publications, presentations, and internal knowledge sharing.
• Provide informal mentorship and technical guidance to junior scientists and collaborators.
• Stay current with advances in AI, computer vision, and medical imaging to continuously elevate technical approaches.

Qualifications

• Required: Doctorate degree in Computer Science, Artificial Intelligence, Machine Learning, Electrical Engineering, Computational Biology, or a related quantitative field with 3 years of related experience.
• Required: Or Master’s degree in related field with 6 years of related industry experience.
• Required: Or Bachelor’s degree in related field with 8 years of related industry experience.
• Preferred: Demonstrated deep technical expertise in developing custom AI/ML models for medical imaging.
• Preferred: Strong experience in digital pathology and/or radiology, including whole-slide images and/or CT, MRI, or PET.
• Preferred: Expertise in foundation model usage, including pre-training, fine-tuning, and domain adaptation for imaging-based tasks.
• Preferred: Advanced knowledge of CNNs, U-Net–based architectures, Vision Transformers; self-supervised, weakly supervised, and few-shot learning; multimodal and representation learning approaches.
• Preferred: Proficiency in Python and deep learning frameworks such as PyTorch and/or TensorFlow.
• Preferred: Ability to communicate complex technical concepts clearly and influence scientific decision-making.
• Preferred: Strong record of scientific contributions (publications, patents, deployed models, or platform capabilities).
• Preferred: Experience setting and implementing technical or scientific strategies for complex AI/ML or computational imaging initiatives.
• Preferred: Strong publication record in AI/ML with applications to drug development, biomarker discovery, patient stratification, or translational research.
• Preferred: Experience working with integrated imaging, clinical, and molecular datasets.
• Preferred: Familiarity with MLOps, scalable training, and model lifecycle management.
• Preferred: Experience with cloud or HPC environments (e.g., AWS, Azure, GCP, SLURM), containerization, and distributed training.
• Preferred: Prior experience leading significant components of cross-functional or external collaboration.

Skills

• Python programming
• PyTorch or TensorFlow
• Medical imaging analysis
• Computer vision techniques (CNNs, U-Net, Vision Transformers)
• Foundation models, self-supervised learning, multimodal learning
• Data integration (imaging with clinical/molecular data)
• Model evaluation, reproducibility, and statistical rigor
• MLOps concepts and scalable training
• Cloud/HPC environments and distributed training
• Communication of complex technical concepts to diverse stakeholders

Education

• Doctorate/Master/Bachelor degrees as listed in Qualifications in related quantitative fields.