We launched recruiters.rip, let your AI apply for you. First month free with code onemonthfree

Software Engineer - Red Team

15 hours agoSalary: 65000 - 145000 GBP / yearLondon, UKHybridEngineeringJobs by The AI Security Institute (AISI)
The AI Security Institute (AISI) logo
The AI Security Institute (AISI)Visit The AI Security Institute (AISI) website

UK government research organisation that evaluates advanced AI risks and develops and tests mitigations to inform governments.

London, United Kingdom

Investors

About The AI Security Institute (AISI)

The AI Security Institute is a research organisation within the UK Department for Science, Innovation and Technology. It tests leading AI systems, conducts technical research on advanced-AI risks and solutions, develops evaluation infrastructure, and informs UK and allied-government policy.

View jobs by The AI Security Institute (AISI)

Skills

Candidate Availability

About the Role

You will join the Red Team as a software engineer, building and maintaining systems that run red-teaming against frontier AI models at scale. You will design systems for large scale automated research with many agents, extend evaluation tooling built on Inspect, and solve performance, reliability and security problems. You will help run pre-release testing, turn research ideas into well-tested maintainable code and support researchers in writing better software.

Requirements

  • Experience writing production quality code at fast pace
  • Python experience
  • Experience building and maintaining systems on AWS or other cloud providers using infrastructure-as-code
  • Strong written and verbal communication skills
  • Experience collaborating with many people including external teams
  • Interest in automating work with agents
  • Baseline personnel security standard (BPSS) clearance and criminal record check

Responsibilities

  • Build and maintain systems for running red-teaming against frontier models at scale
  • Design and implement systems for large scale automated research and engineering with many agents
  • Extend evaluation tooling built on Inspect to new kinds of targets
  • Develop solutions to performance, reliability and security problems from running thousands of agents in parallel
  • Work with Core Technology to share infrastructure
  • Help run pre-release testing of frontier models to tight deadlines
  • Debug problems and ensure results are reliable and reproducible
  • Turn loosely defined research ideas into well-tested, documented, maintainable code
  • Review code and share knowledge with researchers

Benefits

  • Hybrid working
  • Flexibility for occasional remote work abroad
  • Stipends for work-from-home equipment
  • Annual stipends for learning and development
  • Funding for conferences and external collaborations
  • At least 25 days' annual leave
  • 8 public holidays
  • 3 days off for volunteering
  • Extra team-wide breaks
  • Paid parental leave
  • Civil Service pension with up to 14.75% employer contributions
  • Cycling to work, donation and retail/gym discounts

Hiring Process

Initial assessment, initial screening call, technical assessment, behavioural interview, and final interview with senior leadership team members.