Build datasets, evaluations, and scalable data infrastructure for improving frontier AI models on biological and scientific tasks. Analyze model failure modes, design rigorous evaluations, run experiments, and collaborate with AI labs and scientists to translate expert judgment into reliable evaluation criteria. The role combines software engineering, biology, and frontier AI research in an ambiguous, rapidly evolving environment.

Build end-to-end AI agents that automate scientific workflows, including experiment design, data capture, analysis, and reporting. Collaborate directly with biotech customers to identify use cases, gather feedback, develop evaluations, and onboard teams. Engineer across backend and frontend systems, improve agent infrastructure and tooling, and influence Benchling’s AI technical direction and product development practices.