AI Security Engineer — Frontier Cyber Defense

Aisi; Greater London; Hybrid; GBP 90,000 - 120,000; Full time

Benefits offered by this job

Hybrid working

Learning stipends and time off for dev

Conference funding

Job summary

AI Security Institute, based in London, seeks exceptional cybersecurity experts to advance frontier AI security. You’ll design cyber ranges, run pre-deployment tests, and validate novel model behaviors while collaborating with engineers and researchers who value autonomy.

The role offers hybrid working options, exposure to pre-release frontier models, and opportunities to publish and collaborate with national security partners.

Qualifications

• Strong Python skills with automation or security tooling.
• Proven experience in cybersecurity red-teaming: pen tests, cyber ranges, CTF design or tooling.
• Interest in improving AI safety and overseeing frontier AI systems.

Responsibilities

• Design cyber ranges and CTF-style challenges for evaluating AI security.
• Build agentic scaffolding to test frontier models with security tools.
• Design metrics and interpret results of cyber capability evaluations.
• Collaborate to ensure robust, scalable evaluation environments.
• Write reports and share findings with stakeholders.

Skills

Python programming

Cybersecurity red-teaming

Automation scripting

Security tooling

CTF experience

Tools

Penetration testing frameworks

Network packet capture tools

Reverse engineering tools

Virtualisation tooling

Job description

AI Security Institute, based in London, seeks exceptional cybersecurity experts to advance frontier AI security. You’ll design cyber ranges, run pre-deployment tests, and validate novel model behaviors while collaborating with engineers and researchers who value autonomy.

The role offers hybrid working options, exposure to pre-release frontier models, and opportunities to publish and collaborate with national security partners.