Research Scientist/Engineer (Evaluations)

About the role

We develop and run evaluations that help assess the risks posed by scheming AIs. You will get to work with frontier labs like OpenAI, Anthropic, and Google DeepMind and be amongst the first to interact with new models before anyone else. The ideal candidate loves rigorously testing frontier AI models, and enjoys building efficient pipelines and automating them. YOU WILL HAVE THE OPPORTUNITY TO - Run pre-deployment evaluation campaigns on the most capable AI systems in the world. We partner with multiple labs, giving you access to a breadth of models that no single AI lab could offer. You'll be among the first people to interact with new models before anyone else. - Deep dive into AI cognition. Scan through thousands of model transcripts to surface behavioral patterns that no one has ever observed before. These patterns are often deeply surprising and fascinating to study, e.g. the non-standard language and the reward-seeking reasoning described in our anti-scheming paper. - Build new evaluations for frontier risks, from designing novel test environments to scaling them across hundreds of distinct scenarios. - Work directly with frontier AI developers. Share your findings, engage with their feedback, and see your evaluations directly inform deployment decisions for the most capable AI systems in the world. - Automate and improve the evaluation pipeline. We already use automation across building, running, and analyzing evals. Rapid progress in agent capabilities opens up radically new possibilities, and you'll have the freedom to rethink and reshape the pipeline as they emerge. KEY REQUIREMENTS - Software engineering skills: Our entire stack uses Python. We're looking for candidates with strong software engineering experience. Ideally, you have experience shipping and maintaining production Python code, and know how to factor messy problems into clean abstractions that others can use and extend. - Process optimisation: You always try to improve workflows. Pre-deployment evaluations are very fast paced so ideally you love shaving friction off your workflows wherever possible. - Data Analysis & Pattern Recognition: You can extract signal from large, messy datasets. You're comfortable with quantitative analysis and know when qualitative assessment is more appropriate. You can identify anomalies and unexpected model behaviors. - Writing and communication: You succinctly convey qualitative and quantitative findings to a technical and non-technical audience. - AI power-user: You are curious about the capabilities and propensities of frontier AI models. You have experience using different models, know which ones to use for which tasks, when not to use AI, and you always experiment with new AI workflows (Bonus) We are using Inspect as our primary evals framework, and we value experience with it. We want to emphasize that people who feel they don't fulfill all of these characteristics but think they would be a good fit for the position, nonetheless, are strongly encouraged to apply. We believe that excellent candidates can come from a variety of backgrounds and are excited to give you opportunities to shine. We don't require a formal background or industry experience and welcome self-taught candidates. ABOUT APOLLO RESEARCH The rapid rise in AI capabilities offer tremendous opportunities, but also present significant risks.AtApollo Research, we're primarily concerned with risks from Loss of Control, i.e. risks coming from the model itself rather than e.g. humans misusing the AI. We're particularly concerned with deceptive alignment / scheming, a phenomenon where a model appears to be aligned but is, in fact, misaligned and capable of evading human oversight. We work on the detection of scheming (e.g., building evaluations and novel evaluation techniques), the science of scheming (e.g., model organisms and the study of scaling trends), and scheming mitigations (e.g., control). We closely work with multiple frontier AI companies, e.g. to test their models before deployment and collaborate on fundamental research. At Apollo, we aim for a culture that emphasizes truth-seeking, being goal-oriented, giving and receiving constructive feedback, and being friendly and helpful. If you're interested in more details about what it's like working at Apollo, you can find more information here. The current evals team consists of Jérémy Scheurer,Alex Meinke,Bronson Schoen, Felix Hofstätter,Axel Højmark,Teun van der Weij,Alex Lloyd and Mia Hopman.Alex Meinke coordinates the research agenda with guidance from Marius Hobbhahn, though team members lead individual projects. You will mostly work with the evals team as well as our team of software engineers, but you will likely sometimes interact with the governance team to translate technical knowledge into concrete re

Additional Information

Application deadline: We are conducting interviews actively and aim to fill this role as soon as we find someone suitable.

Your Match

How well this role fits your profile.

Company Intel

What employees say

Worked at apolloresearch? Share your experience