Computational Scientist 3, Spatial Omics & Computational Pathology

Roche • South San Francisco, CA

2d • $129,200 - $240,000 • Onsite

About The Position

The Research Pathology Department, an integral part of Genentech’s Research and Early Development Organization (gRED), is dedicated to ensuring that strategies for the treatment of diseases are grounded in accurate analyses of pathogenetic mechanisms. Building upon a strong foundation in digital pathology, the department is at the forefront of advancing spatial omics capabilities, integrating cutting-edge, tissue-based technologies with computational methods to enable high-resolution spatial profiling of biological systems. DPIA-SO (Digital Pathology Image Analysis-Spatial Omics) is a specialized team within Research Pathology focused on collaborative spatial omics computational analysis. We are seeking a highly skilled Computational Scientist to join our team, operating at the intersection of computer vision, advanced machine learning, computational pathology, and spatial biology. The core focus is to develop models, digital pathology infrastructure, and AI pipelines needed to power spatial omics initiatives and scientifically driven projects. We welcome individuals from computational pathology and computer vision backgrounds who possess highly transferrable skills. Candidates with direct spatial omics machine learning expertise represent an ideal fit.

Requirements

• Ph.D in Computational Biology, Computer Science, Machine Learning, Imaging Science, Data Science, or a related highly quantitative field or a Masters Degree in these fields with 3+ years of experience may be considered.
• Demonstrated experience in computer vision, deep learning, or image processing, specifically with tissue-based or high-dimensional imaging data.
• Strong foundation in digital/computational pathology workflows and/or advanced machine learning (e.g., probabilistic modeling, representation learning, generative modeling).
• Deep proficiency in Python software engineering and extensive hands-on experience with modern machine learning frameworks (e.g., PyTorch, TensorFlow, JAX).
• Excellent problem-solving skills with the ability to work independently as a technical lead in a multidisciplinary environment.

Nice To Haves

• Demonstrated experience applying machine learning to single-cell spatial transcriptomics and/or spatial proteomics analysis (e.g., 10X Genomics Xenium, Visium, Lunaphore COMET).
• Hands-on experience with multi-modal data integration, specifically combining spatial transcriptomics, proteomics, and histology datasets.
• Familiarity with the scverse ecosystem (e.g., Scanpy, Squidpy, SpatialData, scVI), computer vision libraries (OpenCV, scikit-image), and modern cloud infrastructure.
• Solid understanding of tissue histology, cell biology, and tumor microenvironments to inform model architecture.
• Experience developing agentic AI systems, LLM-driven autonomous workflows, or advanced AI-oriented tools for complex biological datasets.

Responsibilities

• Serve as the Technical Lead for computer vision, AI/ML, and multiplex imaging projects, architecting pipelines for standard whole-slide images (H&E, IHC) and high-dimensional spatial omics data.
• Design and implement cutting-edge machine learning algorithms, including foundation models and generative architectures, for image segmentation, feature extraction, and predictive modeling.
• Develop advanced multi-modal representation learning frameworks to harmonize disparate modalities—fusing morphological features with molecular data (transcriptomics/proteomics) to uncover spatial niches and cell-cell interactions.
• Engineer scalable imaging data infrastructure for large-scale image storage (e.g., OME-ZARR, OME-TIFF, SpatialData) on HPC and cloud environments.
• Embed biological priors—such as known metabolic pathways or spatial knowledge graphs—directly into the mathematical design of the AI models.
• Collaborate closely with pathologists, wet-lab, and dry-lab researchers to interpret data, visualize results, and contribute to upstream experimental design.

Benefits

• Relocation benefits are available for this posting.
• A discretionary annual bonus may be available based on individual and Company performance.
• This position also qualifies for the benefits detailed at the link provided below.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior