Data Scientist, AI for Biomedical Imaging
Novartis
August 22, 2026
On-site
Cambridge, MA
IT
Role Responsibilities
- Lead AI-enabled image analysis strategies for complex biological imaging workflows, partnering with wet-lab scientists to understand assay needs and explain AI concepts.
- Identify high-impact opportunities for AI and define benchmarking/evaluation strategies for method selection.
- Develop, validate, and deploy robust image analysis algorithms to characterize cellular, organoid, tissue, and other phenotypes in high-throughput/high-content imaging; produce reproducible outputs for drug discovery decisions.
- Drive adoption of advanced imaging AI methods (deep learning, vision foundation models, embedding-based phenotyping, segmentation, classification, multimodal integration) into practical, validated workflows.
- Build scalable, reusable image analysis workflows with data scientists, data engineering, and platform teams; champion best practices.
Essential Requirements
- PhD in CS/AI/ML/biomedical image analysis/computational imaging/data science or related field; 3+ years applied experience in AI for bioimaging/CV.
- Experience developing/validating image analysis algorithms for biological/biomedical/pharmaceutical research (segmentation, feature extraction, phenotypic profiling, object classification, representation learning).
- Practical expertise designing benchmarking and evaluation strategies.
- Ability to work in Linux HPC, cloud, or large-scale data environments with reproducible research, version control, testing, and data provenance.
- Proficiency in Python and deep learning stack (e.g., PyTorch, Hugging Face, Lightning, MONAI) plus tools (scikit-image, OpenCV, napari, Cellpose, StarDist, InstanSeg, OME-Zarr).
- Strong scientific communication with technical and non-technical stakeholders.
Desirable Requirements
- Foundation models/self-supervised/multimodal or embedding-based methods for biological imaging (e.g., DINO, CLIP, SAM).
- Familiarity with drug discovery and phenotypic screening.
- Evidence of reusable/scalable workflows or analysis products integrated into platforms/pipelines/tools.
- Publication/presentations/open-source contributions or technical leadership.
- Familiarity with AI-assisted development/agentic coding (e.g., Claude Code, Copilot).
Benefits & Application Instructions
- Performance-based cash incentive; possible eligibility for annual equity awards (depending on level).
- US-based eligible employees: comprehensive benefits package including health/life/disability and 401(k) with match; generous time off (vacation/personal days/holidays/leaves).
- Lead AI-enabled image analysis strategies for complex biological imaging workflows, partnering with wet-lab scientists to understand assay needs and explain AI concepts.
- Identify high-impact opportunities for AI and define benchmarking/evaluation strategies for method selection.
- Develop, validate, and deploy robust image analysis algorithms to characterize cellular, organoid, tissue, and other phenotypes in high-throughput/high-content imaging; produce reproducible outputs for drug discovery decisions.
- Drive adoption of advanced imaging AI methods (deep learning, vision foundation models, embedding-based phenotyping, segmentation, classification, multimodal integration) into practical, validated workflows.
- Build scalable, reusable image analysis workflows with data scientists, data engineering, and platform teams; champion best practices.
Essential Requirements
- PhD in CS/AI/ML/biomedical image analysis/computational imaging/data science or related field; 3+ years applied experience in AI for bioimaging/CV.
- Experience developing/validating image analysis algorithms for biological/biomedical/pharmaceutical research (segmentation, feature extraction, phenotypic profiling, object classification, representation learning).
- Practical expertise designing benchmarking and evaluation strategies.
- Ability to work in Linux HPC, cloud, or large-scale data environments with reproducible research, version control, testing, and data provenance.
- Proficiency in Python and deep learning stack (e.g., PyTorch, Hugging Face, Lightning, MONAI) plus tools (scikit-image, OpenCV, napari, Cellpose, StarDist, InstanSeg, OME-Zarr).
- Strong scientific communication with technical and non-technical stakeholders.
Desirable Requirements
- Foundation models/self-supervised/multimodal or embedding-based methods for biological imaging (e.g., DINO, CLIP, SAM).
- Familiarity with drug discovery and phenotypic screening.
- Evidence of reusable/scalable workflows or analysis products integrated into platforms/pipelines/tools.
- Publication/presentations/open-source contributions or technical leadership.
- Familiarity with AI-assisted development/agentic coding (e.g., Claude Code, Copilot).
Benefits & Application Instructions
- Performance-based cash incentive; possible eligibility for annual equity awards (depending on level).
- US-based eligible employees: comprehensive benefits package including health/life/disability and 401(k) with match; generous time off (vacation/personal days/holidays/leaves).