Evals Platform Engineer

Apollo Research is at the forefront of AI safety, focusing on evaluating and auditing real-world AI models to address critical challenges like deceptive alignment. As a Platform Engineer, you'll join a dynamic team working on frontier AI evaluation research, with a strong emphasis on security. You'll have broad decision-making authority on the infrastructure stack, designing and maintaining the systems that researchers depend on daily.

The role combines infrastructure expertise with hands-on development, working alongside software engineers and research scientists to ensure scalable, secure, and efficient systems. You'll be part of the Evals team, collaborating with experienced professionals like Rusheb Shah and Andrei Matveiakin, while interacting with the broader research team.

Key projects include building job orchestration systems for LLM evals, implementing secure databases, managing permissions structures, and maintaining secure development environments. The position offers significant autonomy in technical decisions while contributing to crucial AI safety research.

The company culture emphasizes truth-seeking, goal-oriented work, and constructive feedback. Located in London and sharing space with LISA offices, the role offers competitive compensation, flexible working arrangements, and comprehensive benefits. Apollo Research values diversity and provides equal opportunities to all candidates.

This is an excellent opportunity for an experienced infrastructure specialist who wants to make a meaningful impact in AI safety while working with cutting-edge technology and a talented, dedicated team.