Frontier AI Evaluation Scientist

Apollo Research in London or San Francisco is seeking engineers to join a team conducting pre-deployment evaluations of frontier AI systems, collaborating with leading labs to test models before deployment.

You will rigorously analyze model transcripts to surface unexpected behaviors, design new tests, and scale evaluations across hundreds of scenarios, while automating pipelines and sharing findings with developers and stakeholders.

#J-18808-Ljbffr…