Software engineer in red teaming

грейд
middle
формат
гибрид Лондон
язык
Английский B1
опубликована
сегодня

описание

Flexibility for occasional remote work abroad.

The AI Security Institute is the world's largest and best-funded team dedicated to understanding advanced AI risks and translating that knowledge into action. We’re in the heart of the UK government with direct lines to No. 10 (the Prime Minister's office), and we work with frontier developers and governments globally.

задачи

  • Build and maintain systems for running red-teaming against frontier models at scale, such as automated jailbreaking pipelines, alignment auditor agents in Inspect Petri, and attacker agents for control red-teaming
  • Design and implement systems for running large scale automated research and engineering with many agents on AISI’s research platform
  • Extend our evaluation tooling, built on Inspect (AISI's open-source framework for evaluating AI), to new kinds of targets
  • Develop novel solutions to performance, reliability, and security problems that arise from running thousands of agents in parallel
  • Work with Core Technology to share infrastructure and avoid duplicating work
  • Help run pre-release testing of frontier models, often to tight deadlines
  • Debug problems as they come up and make sure results are reliable and reproducible
  • Turn loosely defined research ideas into well-tested, documented and maintainable code
  • Review code, share knowledge and help researchers write better software

требования

  • Writing production quality code at fast pace.
  • Interest in developing ways to automate work with agents, while accounting for their limitations.
  • Designing, shipping, and maintaining complex tech products.
  • Improving technical standards across a team, through mentoring and feedback.
  • Strong written and verbal communication skills.
  • Collaborating effectively with many people, including external teams.
  • Experience working with a world-class research team comprised of both scientists and engineers (e.g. in a top-3 lab).
  • Python experience
  • Experience building and maintaining systems on AWS or other cloud providers using infrastructure-as-code.
  • Будет плюсом: Expertise in Cloud Infrastructure or Dev Ops (AWS, Azure, Kubernetes, Terraform, CDK, Docker, etc.); Cybersecurity expertise; ML Ops (vLLM, agent frameworks, fine-tuning, RAG systems, etc.); Experience with red teaming, adversarial machine learning, or offensive or adversarial security (for example penetration testing or capture-the-flag competitions); Hands-on experience with large language model APIs and the tools around them, such as OpenAI-compatible APIs, agent frameworks or vLLM; Building, running or analysing AI evaluations, ideally with Inspect; Cloud infrastructure and DevOps experience with AWS, Kubernetes, Docker, Terraform or similar; Familiarity with how AI could be misused or could slip out of human control, and how safeguards try to prevent this

условия

  • Incredibly talented, mission-driven and supportive colleagues.
  • Direct influence on how frontier AI is governed and deployed globally.
  • Work with the Prime Minister’s AI Advisor and leading AI companies.
  • Opportunity to shape the first & best-resourced public-interest research team focused on AI security.
  • Pre-release access to multiple frontier models and ample compute.
  • Extensive operational support so you can focus on research and ship quickly.
  • Work with experts across national security, policy, AI research and adjacent sciences.
  • If you’re talented and driven, you’ll own important problems early.
  • 5 days off and annual stipends for learning and development, and funding for conferences and external collaborations.
  • Freedom to pursue research bets without product pressure.
  • Opportunities to publish and collaborate externally.
  • Modern central London office, or where applicable, option to work in similar government offices in Birmingham, Cardiff, Darlington, Edinburgh, Salford or Bristol.
  • Hybrid working, flexibility for occasional remote work abroad and stipends for work-from-home equipment.
  • At least 25 days’ annual leave, 8 public holidays, extra team-wide breaks and 3 days off for volunteering.
  • Generous paid parental leave (36 weeks of UK statutory leave shared between parents + 3 extra paid weeks + option for additional unpaid time).
  • Excellent pension provision, with a choice between Partnership (no employee contributions required and up to 14.75% employer contributions) or Alpha, a valuable defined benefit scheme that provides a guaranteed, inflation-linked retirement income for life.
  • Discounts and benefits for cycling to work, donations and retail/gyms.
  • Annual salary is benchmarked to role scope and relevant experience. Most offers land between £65,000 and £145,000 made up of a base salary plus a technical allowance (take-home salary = base + technical allowance).
  • This role sits outside of the DDaT pay framework given the scope of this role requires in depth technical expertise in frontier AI safety, robustness and advanced AI architectures.
  • Level 3: £65,000–£75,000 (Base £40,850 + Technical Allowance £24,150–£34,150)
  • Level 4: £85,000–£95,000 (Base £48,540 + Technical Allowance £36,460–£46,460)
  • Level 5: £105,000–£115,000 (Base £63,165 + Technical Allowance £41,835–£51,835)
  • Level 6: £125,000–£135,000 (Base £76,475 + Technical Allowance £48,525–£58,525)
  • Level 7: £145,000 (Base £76,475 + Technical Allowance £68,525)

навыки

похожие вакансии

статьи для разработчиков