Note: The job is a remote job and is open to candidates in USA. GSK is a global biopharma company focused on discovering and delivering vaccines and medicines to improve patient health. The Principal Scientist will lead translational AI and machine learning research integrating multimodal oncology data, including functional genomics, spatial omics, histopathology, and real-world clinical data. The role will develop and validate models, generate actionable oncological hypotheses, support clinical pipeline decisions, and advance reproducible research practices.
Responsibilities
- Own the pipeline and develop advanced ML architectures to integrate complex multimodal datasets, including single-cell, spatial omics, histopathology, functional genomics, and real-world clinical data
- Partner closely with wet-lab scientists, clinicians, and pathologists to validate machine learning models, including in-silico perturbations within the tumor microenvironment against ground-truth data (counterfactual validation)
- Develop approaches to extract interpretable features from models to generate testable oncological hypotheses and link insights to clinical pipeline decisions such as asset prioritization and patient subpopulation selection
- Contribute clean, reproducible tooling to cross-team frameworks. We enforce good engineering practices in our research—utilizing code architecture planning, clean code and automated testing to build trustworthy, reusable code
- Maintain cutting edge knowledge of advancements, share with the team and maintain our team as a thought leader through publications in high-impact venues and engaging with the broader community
Skills
- PhD (or equivalent experience) in a quantitative field (Applied ML, Computer Science, Physics, Systems/Computational Biology, or equivalent) with 1+ years of industry or productive post-doctoral academic experience
- Experience with deeply embedded in cancer / computational biology, with a strong understanding of tumor microenvironment dynamics and high dimensional datasets
- Experience with analytical and modelling skills, including expertise in statistical and machine learning approaches
- Experience with the analysis of single cell omics data
- Experience in one or more of the following: statistical modelling of functional genomics screening datasets (e.g., bulk CRISPR screens, Perturb-seq) or spatial omics
- Experience in Python and deep learning frameworks (PyTorch) for data processing and machine learning model development, with a strong grasp of software engineering fundamentals (e.g., version control, modular design, CI/CD)
- Experience with multi-modal integration, including spatial transcriptomics/proteomics, histopathology, and single cell omics data
- Experience working with longitudinal clinical health record trajectory data
- Familiarity with R for specialized statistical modelling
- Experience with causal inference and individual treatment effect modelling
- Experience with AI agent-driven workflows and coding tools
- Experience with generative deep learning approaches, including flow matching, diffusion and causal transformer models
- Excellent written and oral communication skills, with a proven ability to present complex computational concepts to technical and non-technical stakeholders
Benefits
- Hybrid work arrangement balancing on-site collaboration with focused remote work.
- This position offers an annual bonus.
- Eligibility to participate in our share based long term incentive program, dependent on the level of the role.
- Health care benefits for employee and family.
- Other insurance benefits for employee and family.
- Retirement benefits.
- Paid holidays.
- Vacation.
- Paid caregiver/parental and medical leave.
Company Overview
Company H1B Sponsorship