Frontier AI Safety Research Engineer

AI Security Institute seeks researchers and engineers to design and run ML experiments on frontier models, building adversarial attacks and evaluating monitors. You will write rigorous safety arguments, perform threat modelling, and contribute to reports guiding government and industry actions.

This role emphasizes autonomy, collaboration, and high-velocity research in a leadership-driven environment. The team works on control measures, sandboxing, and evaluation pipelines, with access to