Senior Scientist, Machine Learning (Biologics Design)

Description de l'emploi

Gilead’s Research Data Sciences is seeking a Senior Scientist to develop and apply machine learning methods for the design and optimization of large molecules, including antibodies, multispecifics, and other complex formats. This role sits at the intersection of machine learning, structural biophysics, and protein therapeutics, with direct impact on lead optimization and pipeline programs.

You will build predictive and generative models that guide sequence and structure design, integrate diverse experimental and structural datasets, and work in close partnership with experimental teams. A key emphasis is data-efficient learning, using limited and noisy experimental data to make high-confidence design decisions.

Key Responsibilities

• Develop and apply ML models for biologics design, including sequence-to-function, structure-aware, and multi-objective models that support lead optimization decisions
• Implement data-efficient modeling strategies (e.g., active learning, Bayesian optimization, experimental design) to prioritize designs and guide iterative experimentation
• Apply and extend modern deep learning approaches relevant to biologics, including protein language models, geometric deep learning, and generative methods (e.g., diffusion, inverse folding, ProteinMPNN-style approaches)
• Perform structure-based modeling and analysis of antibodies and multispecifics.
• Partner closely with protein therapeutics, structural biology, assay, and engineering teams to translate computational results into experimental decisions

Required Qualifications

• PhD in Computational Biology, Computer Science, Mathematics, Physics, Chemistry, Bioengineering, or a related quantitative discipline, and 2+ years of experience
• Strong proficiency in Python and deep learning frameworks such as PyTorch (and/or JAX), plus standard scientific libraries (NumPy, pandas, etc.)
• Demonstrated experience architecting, training, and evaluating deep learning models, such as representation learning, multimodal learning, geometric deep learning, or generative modeling
• Solid understanding of protein structure, antibody architecture, and biophysical principles relevant to large-molecule therapeutics
• Demonstrated research productivity (e.g., first-author publications), and ability to communicate clearly to diverse audiences

Preferred Qualifications

• Experience with molecular modeling or simulations (e.g., Amber, OpenMM, Rosetta, CHARMM, coarse-grained or multi-scale methods)
• Experience developing production-grade ML tooling: experiment tracking, model registries, CI/testing, containerization, workflow orchestration
• Prior industry experience in biologics discovery, protein engineering, or therapeutic protein development