Senior Machine Learning Scientist I, Model-Driven Optimization

The Role:  Generate:Biomedicines is seeking a creative, rigorous, and execution-oriented machine learning scientist to join our Model-Driven Design team. This role will focus on building the ML methods, data strategies, and closed-loop systems that determine what we design, build, test, and learn from next. The Model-Driven Design team works at the interface of machine learning, protein design, engineering, and experimental science. We develop and apply models and quantitative frameworks that help Generate discover and optimize therapeutic proteins. In this role, you will help advance the technical foundation of our lab-in-the-loop protein optimization platform, with a focus on sequential decision-making, experimental design, property modeling, and scalable design systems. We are looking for someone who can serve as a technical leader and hands-on individual contributor, driving complex, high-impact work from problem framing through implementation, deployment, and experimental impact. The ideal candidate combines depth in probabilistic machine learning, Bayesian optimization, active learning, or related approaches with the practical judgment and engineering discipline to turn technical ideas into reliable systems that drive impact. You will partner closely with protein designers, wet-lab scientists, ML scientists, and engineers to build durable capabilities that accelerate therapeutic discovery. This role is part of a highly collaborative team environment that balances in-person collaboration with hybrid flexibility based out of our Somerville, MA office.  Here's how you will contribute: Develop new machine learning methods and systems for lab-in-the-loop protein optimization, including property models and multi-objective optimization strategies for therapeutic protein design. Shape data-generation and data-use strategies that make experimental campaigns maximally informative for model improvement, therapeutic optimization, and future design cycles. Build and apply LLM-enabled and agentic workflows that help scientists explore design hypotheses, connect models to data and experiments, and accelerate iterative learning. Design, implement, test, and maintain production-quality ML models, software components, and data workflows, with attention to reliability, reproducibility, observability, and computational efficiency. Partner with ML engineering and software teams to integrate these components into robust, scalable platform capabilities, with clear ownership across team boundaries. Collaborate closely with protein designers and wet-lab scientists to ensure models and optimization systems are grounded in experimental reality and deliver measurable impact. Identify important technical gaps, develop proposals, define milestones, align stakeholders, and help set technical direction across cross-functional programs. Communicate clearly across disciplines and help raise technical standards across ML, engineering, protein design, and experimental teams. The Ideal Candidate will have: PhD in machine learning, computational biology, computer science, applied mathematics, engineering, or a related quantitative field. Strong practical experience with probabilistic machine learning, Bayesian optimization, active learning, experimental design, or related approaches for sequential decision-making under uncertainty. Experience developing machine learning methods or systems for biological, biomedical, or experimental scientific data, with an ability to reason about noisy assays, sparse labels, experimental bias, and data-generation strategy. Demonstrated ability to translate ML ideas into systems, tools, or workflows that affect real scientific, experimental, or product decisions. Strong Python skills and experience with modern ML frameworks such as PyTorch, JAX, or similar tools. Strong systems thinking and ability to design technical interfaces, reason about system tradeoffs, and partner with engineering teams to build scalable, maintainable ML infrastructure. Excellent communication skills and ability to bridge ML, engineering, protein design, and experimental stakeholders. Pragmatic, collaborative working style, with the ability to bring structure to open-ended problems and balance scientific rigor with execution in fast-moving, cross-functional environments. Nice to have Experience in protein design, protein engineering, antibody engineering, biologics discovery, or drug development. Experience partnering with experimental teams on design-build-test-learn cycles, high-throughput screening, directed evolution, pooled libraries, or model-guided experimental campaigns. Experience with multi-objective optimization, uncertainty calibration, model-guided library design, or experimental campaign planning. Experience developing and applying deep learning models, including transformer-based architectures Experience building or applying LLM agents, scientific copilots, or agentic syste