The posting, in Generate Biomedicines's own words
archived Oct 21, 2026The Role:
We are seeking an inventive Computational Protein Design Scientist to join our team working at the interface of computational protein design, machine learning, and experimental biology. In this role, you will use your computational expertise and in-depth biochemical understanding of proteins to drive computational protein design strategy for therapeutic programs while helping develop reusable design capabilities to unlock novel design challenges. The role interfaces between our machine learning and experimental platform teams, working closely to efficiently integrate AI-designed proteins and experimental data. You will combine hands-on protein design with computational experimentation and close partnership across program, experimental, and computational teams. You will translate therapeutic hypotheses into design and data-generation plans, execute protein design campaigns, interpret results, and help guide decision-making in fast-paced cross-functional teams. You will also feed back application learning to drive platform capability development by identifying recurring needs, successful approaches, and failure modes. This work will help shape new design methods, workflows, and scientific requirements that make program execution more efficient, reproducible, and scalable.
Here's how you will contribute:
Drive therapeutic protein design. Leverage novel computational tools, biochemistry, and structural biology knowledge to translate therapeutic hypotheses and program needs into end-to-end protein design campaigns, data strategies, and technical milestones, highlighting and navigating key uncertainties. Connect application learning to capability development. Identify recurring capability gaps, successful approaches, and failure modes, and translate these insights into improved methods, workflows, and priorities. Connect design and experimental learning. Serve as an effective interface between computational model development and experimental validation, feeding biological learnings back to computational collaborators and vice versa. Use biochemical background to shape the most informative experiments involving complex assay readouts, signal and noise properties, and experiment design tradeoffs. Develop and apply agent-enabled computational workflows. Contribute directly to building workflows that connect scientific information, protein-design models, data, and tools, and apply them to program questions and design decisions. Evaluate their outputs and reliability in the context of the decisions they support. Translate results into program decisions. Develop and communicate clear, data-driven design recommendations that connect molecular observations to program context and next-step decisions, and document assumptions, results, and limitations so the work is understandable and reusable.