Principal Scientist, Computational Sciences - Protein Structure Prediction and Design

Bristol Myers Squibb is seeking an experienced, creative, and highly collaborative scientist to join our Biotherapeutics Computational Design team. In this role, you will build and deploy cutting-edge machine learning and structure-based methods to accelerate biologics discovery across the preclinical portfolio (antibodies, multispecifics, ADCs, and novel scaffolds) and play a pivotal role in scaling an agentic antibody design platform from prototype to a core engine of research innovation.

This position offers a unique opportunity to work at the intersection of machine learning, structural biology, and drug discovery. You will leverage large-scale proprietary and public datasets and collaborate with cross-functional teams of scientists to address complex challenges in biologics discovery and engineering, directly supporting and accelerating the development of next-generation therapies.

Key Responsibilities

• Develop and scale antibody design capabilities from prototype to application: advance agentic antibody design approaches into robust, reusable workflows that support preclinical discovery.
• Build and apply state-of-the-art models for biologics design: protein structure modeling, binder design, affinity/specificity prediction, and developability property prediction using internal and external datasets.
• Deliver reliable, production-ready research tools: own end-to-end development of computational pipelines with strong emphasis on reproducibility, benchmarking, and maintainable, well-documented code.
• Lead through influence: partner with computational and wet-lab teams to prioritize capabilities, translate insights into actionable decisions, and communicate clearly to technical and non-technical audiences.

Required Qualifications

• Ph.D. in structural bioinformatics, computational biology, computer science, engineering, physics, or a related discipline, with 4+ years of relevant industry or academic experience
• Expertise in modern machine learning approaches (e.g., transformers and diffusion/flow-based generative models) and strong fundamentals in classical machine learning
• Experience developing and evaluating predictive models, including model assessment, benchmarking, and experimental design
• Hands-on experience with protein modeling, including state-of-the-art methods for protein structure prediction and generative protein design
• Experience developing or applying agentic AI frameworks to build applications that automate and accelerate research workflows
• Strong Python skills and commitment to reproducible research and high-quality scientific software
• Ability to identify high-impact problems and work independently to drive solutions through implementation and evaluation
• Ability to collaborate across disciplines and communicate technical findings clearly

Preferred Qualifications

• Experience with physics-based modeling (e.g., molecular dynamics, free energy perturbation) or closed-loop optimization (e.g., Bayesian optimization, active learning)
• Background knowledge in biochemistry, protein engineering, or related experimental disciplines

Compensation (San Diego - CA): $156,890 - $190,117 (additional locations: Brisbane, CA and Cambridge Crossing, MA at $166,770 - $202,086). Additional incentive cash and stock opportunities may be available.