Return to jobs
Exciting opportunity to transform AI research into practical tools at Apollo Research
Overview
Salary
£100000 - £180000
Location
London, UK
Expires
Expires at anytime
Organisation Summary
Apollo Research is at the forefront of making AI research practical by focusing on monitoring coding agents for safety and security failures as part of their mission. This is an amazing chance to be part of a mission that mitigates risks from AI.
Role Summary
- Join as an Applied Researcher to collaborate closely with monitoring and software engineers.
- Create systems to collect and catalogue coding agent failure modes.
- Design experiments, build and maintain evaluation frameworks.
- Contribute in monitor design and optimization, develop monitoring prompts and experiment with reasoning strategies.
Role Requirements
- At least 2 years of experience in empirical research with AI systems.
- Excellent Python coding skills.
- Familiarity with AI safety concepts and computer security.
- Experience in designing and running experiments, analyzing results, and iterating based on findings.
Application Process Details
- Fill in the application form and submit your CV through our website.
- Applications are reviewed on a rolling basis, early submissions are encouraged.
Join Apollo Research and become a pivotal part of our new AGI safety monitoring team. Our mission is to transform complex AI research into practical tools that mitigate risks from AI, focusing on monitoring AI coding agents for safety and security failures.
The role: As an Applied Researcher, you will collaborate closely with our CEO, monitoring engineers, and software engineers from the Evals team. Your main responsibilities will include researching and developing systems to collect and catalog coding agent failure modes, designing experiments to test monitor effectiveness, and building and maintaining evaluation frameworks. You will also be involved in monitor design and optimization, developing a comprehensive library of monitoring prompts, and experimenting with different reasoning strategies to enhance monitor reliability.
Job requirements: We are looking for candidates with at least 2 years of experience in empirical research with large language models or AI systems, strong Python programming skills, and familiarity with AI safety concepts and computer security. Experience in designing and running experiments, analyzing results, and iterating based on findings is crucial.
Benefits: We offer a competitive salary range of 100k - 180k GBP, flexible work hours, unlimited vacation and sick leave, and a professional development budget. Meals are provided on workdays, and we support paid work trips and relevant conferences.
To apply, please complete the application form with your CV on our website. We review applications on a rolling basis and encourage early submissions. For more details about our team and work culture, visit our website.
