Research Scientist Control

Research Scientist Control Overview

Join Apollo Research as a Research Scientist in Control, where you will be part of a pioneering team focused on enhancing AI safety through empirical research. This role involves transforming theoretical AI control research into practical tools aimed at mitigating risks associated with AI systems. You will collaborate closely with the CEO and a dedicated team of researchers and engineers to develop innovative monitoring solutions.

Key Responsibilities

• Design and implement control protocols for AI systems, testing them on real-world production systems.
• Conduct experiments to evaluate the effectiveness of monitoring across various failure modes and agent behaviors.
• Develop and maintain evaluation frameworks to track progress in monitoring capabilities.
• Create and manage high-quality datasets for training and testing monitoring systems.
• Iterate on monitoring strategies based on empirical findings, balancing detection accuracy with computational efficiency.
• Contribute to Apollo’s shared failure mode library by documenting newly identified coding agent failure modes during the design and evaluation phases.
• Stay updated with the latest research on AI safety, agent failures, and detection methodologies.
• Research coding security and safety vulnerabilities to enhance monitoring systems.
• Fine-tune open-source models to develop efficient monitors suitable for high-volume production environments.
• Design and implement autonomous monitoring systems that analyze logs to detect both known and new failure modes.
• Integrate findings from red-teaming campaigns into the design and evaluation of monitoring systems.

Job Requirements

• A minimum of 2 years of experience in empirical research involving large language models or AI systems.
• Strong background in AI coding agents, including extensive use and comparison of leading coding agents or experience in designing/developing coding agents.
• Experience with LLM-as-a-judge setups or broader AI monitoring practices.
• Proficiency in designing and executing experiments, analyzing results, and iterating based on empirical data.
• Strong programming skills in Python.
• Ability to work independently on open-ended research challenges.

Benefits

• Competitive salary and equity options.
• Salary range: $204,000 – $385,000 in San Francisco; £136,000 – £258,000 in London.
• Flexible working hours and schedules.
• Unlimited vacation and sick leave.
• Up to 6 months of paid parental leave.
• Comprehensive health, dental, and vision insurance.
• Retirement savings plan with competitive employer matching.
• Provision of lunch, dinner, and snacks on workdays.
• Paid work-related trips, including retreats and conferences.
• Annual professional development budget of $1,000.
• Relocation support and assistance with visa fees if necessary.

This is a full-time, onsite position based in either London or San Francisco, with flexible working hours and arrangements available. We offer visa sponsorship for qualified candidates, although it is not guaranteed for every role.

How to Apply

To apply, please complete the application form and submit your CV. A cover letter is not required, but you may include links to relevant work samples. The interview process includes a screening interview, a take-home test, three technical interviews, and a final interview with the CEO. We encourage candidates to prepare by building simple monitors for coding agents.

Degree Requirement: Not Specified

Visa Sponsorship Promising