Research Scientist - Science of Evaluations

Job Description

The AI Security Institute is seeking a Research Scientist to join its Science of Evaluations team. This team focuses on measuring existing frontier AI system capabilities and predicting the capabilities of a system before evaluation. The role involves conducting applied and foundational research to develop rigorous scientific techniques for capability measurement.

The Research Scientist will collaborate with researchers and engineers, propose improvements to evaluations, and develop techniques to maximize results. They will also conduct research on predictive evaluations, focusing on complex, long-horizon agent tasks. This role offers the opportunity to contribute to the vision and strategy of the Science of Evaluations team.

Responsibilities Include:

• Running internal red teaming exercises.
• Conducting in-depth analysis of evaluations methodology and results.
• Improving the statistical analysis of evaluations results.
• Developing and implementing internal best-practices and protocols for evaluations and testing exercises.
• Conducting research on predictive evaluations.
• Writing and editing scientific reports.

Requirements Include:

• Experience working within a world-leading team in machine learning or a related field.
• Strong track-record of academic excellence.
• Comprehensive understanding of large language models.
• Broad experience in empirical research methodologies and statistical analysis.
• Deep care about methodological and statistical rigor.
• Experience with data visualization and presentation.
• Proven track record of excellent scientific writing and communication.
• Motivated to conduct technical research with an emphasis on direct policy impact rather than exploring novel ideas.
• Ability to work autonomously and in a self-directed way with high agency.

What the Role Offers:

• Range of salaries available, from £65,000 to £145,000, based on seniority and experience.
• Flexible working arrangements.
• Minimum of 25 days of paid annual leave, increasing with service.
• Extensive learning and professional development opportunities.
• Access to retail, travel, and lifestyle employee discounts.
• Hybrid working policy.