Misuse Red Team – Research Engineer/Research Scientist

Responsibilities

We are representing a confidential AI security organisation that requires research engineers and research scientists to join their misuse-focused research team. In this role, you will focus on safeguarding advanced AI systems from abuse and developing cutting-edge security measures.

Your responsibilities will include:

• Designing, building, and evaluating automated methods to attack and assess AI safeguards, including developing optimisation approaches and LLM-based attacking systems
• Creating benchmarks for asynchronous monitoring to detect signs of system misuse and adversarial development across multiple model interactions
• Researching novel attack vectors and defensive measures for data poisoning, backdoors, and other vulnerabilities in large language models
• Conducting adversarial testing of frontier AI system protections and producing actionable, impactful reports for safeguard developers
• Collaborating with frontier AI companies and government stakeholders to share findings that inform deployment and policy decisions
• Publishing research at top-tier machine learning conferences and contributing to the broader AI security community

Research Scientists typically lead technical direction—formulating questions, designing experiments, and driving conclusions. Research Engineers focus on execution—building scalable systems, ensuring reliability, and maintaining reproducibility. Our client supports work that spans or alternates between these functions.

What We're Looking For

The ideal candidate will have deep expertise in AI security, adversarial robustness, or frontier LLM research. We're seeking individuals at various levels from junior to principal research scientist.

Core qualifications include:

• Hands-on research experience with large language models (training, fine-tuning, evaluation, or safety research)
• Demonstrated track record of peer-reviewed publications in top-tier ML conferences or journals
• Ability to write clean, documented research code for machine learning experiments using frameworks like PyTorch or evaluation tools like Inspect
• Strong sense of mission and personal responsibility for outcomes
• Capability to generate original research ideas while collaborating effectively with team members

Highly valued experience:

• Background in adversarial robustness, AI security, or red teaming against any type of system
• Experience with AI alignment or AI control methodologies
• Extensive production-quality coding experience
• Track record of mentoring and providing constructive feedback to improve team performance
• Experience designing, shipping, and maintaining complex technical products

Candidates should be comfortable working in London and eligible for UK security clearance (baseline personnel security standard with preference for counter-terrorist check clearance). The role offers hybrid working arrangements.

How We Work With You

Our talent team at Get A Job.ai specialises in placing exceptional AI researchers and engineers with organisations at the forefront of AI security. When you apply through our platform:

• A dedicated recruiter will review your application and conduct an initial screening to understand your background and preferences
• We will present qualified candidates to our client with your permission
• The interview process typically includes technical proficiency tests, research interviews, technical assessments, behavioural interviews, and conversations with senior leadership
• We support you throughout the entire process, from initial screening through offer negotiation
• Please do not contact the client directly—all communication should flow through Get A Job.ai

This confidential opportunity offers competitive compensation benchmarked to role scope and experience, with annual salaries typically ranging between £65,000 and £145,000 (comprised of base salary plus technical allowance, with an additional 28.97% employer pension contribution on the base).

Pay

Pay details will be discussed with qualified candidates during the screening process. Compensation is competitive and commensurate with experience in frontier AI research.

Equal Employment Opportunity

Get A Job.ai is committed to equal employment opportunity. We celebrate diversity and are dedicated to creating an inclusive environment for all employees and candidates.