Postdoctoral Research Fellow - Agentic AI for Spatial Modeling of Disease Microenvironments

We are seeking a highly motivated Postdoctoral Research Fellow to develop next-generation AI methods for modeling disease microenvironments using spatial omics, digital pathology, and language-model–based cell representations. This is a joint role spanning Oncology R&D (ORD) and Inflammation & Immunology R&D (I&I), designed to accelerate cross-portfolio discovery through integrative computational modeling of tissue architecture, immune engagement, and disease biology.  
 

The successful candidate will work on an interdisciplinary project aimed at understanding how cellular states, spatial interfaces, and tissue architecture jointly shape therapeutic response in human disease. The postdoc will help build an agentic AI platform that integrates cell-state language models, boundary-resolved spatial profiling, and digital pathology foundation models to generate interpretable, mechanistically grounded insights from multimodal tissue data.  
 

This role offers the opportunity to contribute to both oncology and immunology/inflammation research programs, with a strong emphasis on translational impact, biomarker discovery, reverse translation, and human disease stratification.  

Role Responsibilities
The project is centered on three major scientific ideas:

• Immune activation and suppression may be spatially restricted to defined tissue interface zones.

• Disease transcriptional programs may regulate immune cell biodistribution, accessibility, and functional contact probability.

• Fibroblast- and myeloid-driven suppressive niches may constrain effector-cell distribution and function.  
   

To address these questions, the postdoc will help develop and apply methods that combine:

• Spatial transcriptomics / spatial omics.

• Digital pathology foundation models.

• Cell-state language models, including scGPT, cell2sentence, etc.

• Heterogeneity profiling based on spatial biology and/or digital pathology.

• Agentic AI workflows for iterative multimodal reasoning and analysis.

• Develop computational methods for spatial modeling of tumor–immune and pathogenic tissue–immune interactions using multimodal datasets.

• Build and evaluate AI/ML workflows that integrate spatial omics, histopathology, and clinical outcome data.

• Advance cell-state representation learning using language-model–based approaches for single-cell and spatial biology.

• Apply and extend boundary-resolved profiling methods to quantify immune–disease interactions in spatial contexts.

• Fine-tune and adapt digital pathology foundation models using internal histopathology datasets for biomarker discovery, reverse translation, and patient stratification.

• Contribute to the design of an agentized AI platform for scalable analysis and reasoning over multimodal biomedical data.

• Collaborate closely with scientists across Oncology R&D and Inflammation & Immunology R&D, including computational, translational, pathology, and biology stakeholders.

• Present findings internally and externally, prepare manuscripts, and support the generation of new project ideas and translational hypotheses.
   

Reporting Structure

This postdoctoral position will be jointly monitored and mentored by investigators from both Oncology R&D and Inflammation & Immunology R&D. The postdoc will operate in a highly collaborative matrix environment and will be expected to engage with scientific mentors and stakeholders across both therapeutic areas.  

Required Qualifications

• PhD in Computational Biology, Bioinformatics, Computer Science, Biomedical Engineering, Systems Biology, Statistics, Machine Learning, or a related quantitative discipline.

• Hands-on experience in one or more of the following areas:

  • Spatial transcriptomics and/or single-cell omics

  • Computational pathology and/or digital pathology

  • Large Language Models, Agentic AI, and their applications on computational biology.

• Less than 2 years post-degree experience

• Willingness to make a minimum 2-year commitment.

• Provide two letters of recommendation

• Demonstrated record of scientific accomplishment, evidenced by peer-reviewed scientific publications and/or conference presentations, including at least one first-author publication.

• Proficiency in Python and modern scientific computing and machine learning frameworks.

• Ability to work independently while collaborating effectively within a multidisciplinary research team.

• Strong written and verbal communication skills, with th