Co-op, Agentic and Generative Protein Design

Lila Sciences • San Francisco, MA

2d

About The Position

Lila Sciences is hiring a co-op to work on AI methods for therapeutic protein design, in close collaboration with the wet lab. The position may focus on generative models and protein-design pipelines, AI agents for protein-design campaigns, or a combination of both. The co-op will own a well-scoped project and deliver code and analyses that inform real design decisions and scientific outcomes.

One possible track involves building, benchmarking, and improving methods for designing proteins such as antibodies and minibinders, with candidate designs tested in the lab. The other involves building and evaluating agents that plan and execute design campaigns, including selecting targets, epitopes, methods, and settings; launching design and filtering runs; diagnosing failures; adapting subsequent rounds; and reasoning over biological literature alongside outputs from specialized models.

Lila Sciences is building Scientific Superintelligence™ by combining AI models with proprietary AI Science Factory™ instruments to create systems that can autonomously execute the scientific method across medicine, materials, and energy.

Requirements

• Strong Python skills and a computational science background.
• Experience with coding agents.
• Experience with machine learning for molecular design.

Nice To Haves

• Experience customizing agents, building agent harnesses, or using agents for scientific applications.
• Strong technical understanding of and experience with generative models.

Responsibilities

• Develop, evaluate, and apply agentic systems for protein design and/or deep generative models for protein design.
• Build, benchmark, and improve protein-design methods and pipelines, with designs tested in the wet lab.
• Build and evaluate agents that plan protein-design campaigns, including target and epitope selection and choosing methods and settings.
• Use agents to execute design campaigns by launching design and filtering runs, diagnosing failures, and adapting subsequent rounds.
• Enable agents to reason over biological priors from literature alongside outputs of domain-specific models.
• Own a well-scoped project and deliver code and analyses that support design decisions and scientific outcomes.

What This Job Offers

• Job Type: Full-time
• Career Level: Intern
• Education Level: No Education Listed