Principal Scientist – AI/ML – Predictive modeling – Design and optimization for therapeutic antibodies
| Company | Bristol Myers Squibb |
|---|---|
| Location | Cambridge, MA, USA |
| Salary | $Not Provided – $Not Provided |
| Type | Full-Time |
| Degrees | PhD |
| Experience Level | Senior, Expert or higher |
Requirements
- Ph.D. in computer science, machine learning, statistics, computational biology, or related fields with 6+ years of relevant experience in industry or academia.
- Demonstrated expertise in multi-objective optimization, Bayesian methods, reinforcement learning, or active learning applied to molecular design or related scientific domains.
- Strong background in deep learning, including transformer architectures, variational autoencoders, diffusion models, or GNNs.
- Fluency in Python and modern ML libraries (e.g., PyTorch, TensorFlow, JAX), with excellent software engineering practices (Git, CI/CD, containerization, and reproducible research standards).
- Experience working with large-scale biological datasets, including sequences, assay measurements, and structural data.
- Track record of innovation and technical leadership in complex, multidisciplinary projects.
- Strong written and oral communication skills, with an ability to explain complex models to cross-functional teams and translate business needs into ML solutions.
Responsibilities
- Design, develop, and deploy multi-objective optimization frameworks that guide antibody candidate selection and optimization.
- Build and maintain predictive models that estimate drug-like attributes (e.g., binding affinity, expression, developability, immunogenicity, viscosity) from high-throughput assay and sequence data.
- Collaborate with project teams to formulate optimization goals and define interpretable, model-informed design strategies.
- Lead the application of cutting-edge ML methods, including Bayesian optimization, active learning, transformers, and diffusion models, to antibody design problems.
- Architect pipelines that integrate real-time experimental feedback into active learning or design-make-test-learn loops.
- Translate high-impact innovations into robust tools and workflows deployable across therapeutic programs.
- Contribute to scientific strategy, internal capability building, and external visibility through reports, seminars, meetings, and collaborations.
Preferred Qualifications
- Knowledge in protein biochemistry or antibody engineering is a strong plus.
- Familiarity with high-performance computing and cloud-based infrastructure is desirable.