Evals Platform Engineer

Apollo Research is at the forefront of AI safety, focusing on evaluating and auditing AI systems to prevent deceptive alignment - where models appear aligned but may actually evade human oversight. As their Evals Platform Engineer, you'll be responsible for building and maintaining the secure infrastructure that powers their frontier AI evaluations. This role combines DevOps expertise with AI safety, requiring strong skills in cloud infrastructure, security, and software engineering.

The position offers an opportunity to work with a dedicated team of researchers and engineers in London, alongside the London Initiative for Safe AI (LISA). You'll be instrumental in accelerating research by building robust backend infrastructure and researcher-facing interfaces, while maintaining high security standards for sensitive data handling.

The role demands expertise in AWS, Kubernetes, and Infrastructure as Code, with a focus on Python development. You'll have significant autonomy in technical decision-making while collaborating with a team that values truth-seeking and constructive feedback. The company offers comprehensive benefits including unlimited vacation, provided meals, and professional development opportunities.

This is an ideal position for someone passionate about AI safety who wants to contribute to crucial research through infrastructure development. The role combines technical challenges with meaningful impact in ensuring the safe development of AI systems. With visa sponsorship available and potential relocation support through AI Futures Grants, this position is accessible to international candidates looking to join London's growing AI safety community.