AI Security Researcher — Red Team for Regulated AI

CommonAI C.I.C.; Cambridge; On-site; GBP 65,000 - 90,000; Full time

Benefits offered by this job

Collaborative work environment

High impact role

Competitive salary package and pension

Professional development opportunities

Networking with tech sector and acadêm

Job summary

CommonAI CIC in Cambridge seeks an AI Security Researcher with an attacker mindset to test and secure AI systems used in regulated environments such as financial services. You will track relevant research, assess decisions from an attacker’s view, and help develop interpretable models and robust defenses.

You will threat-model AI systems, conduct red-team testing within scope, and share findings through clear briefings.

Qualifications

• Proven experience in security research, offensive security or adversarial machine learning (red-teaming, vulnerability research, attacks on ML models).
• Strong understanding of how modern AI systems work, including training, deployment and serving of large language models.
• Experience with attacks on AI systems (jailbreaks, prompt injection, data poisoning, backdoors, model extraction).
• Excellent Python skills and ability to build tools and experiments.
• Strong written communication to convey complex findings to technical and non-technical readers.

Responsibilities

• Track key research and developments relevant to the High Assurance programme, including attacks on AI models and defenses.
• Report on implications for the programme through written briefings and presentations.
• Review technical decisions from an attacker’s point of view and propose stronger alternatives.
• Threat-model AI systems used in regulated settings, covering weights, data, inference serving and tooling.
• Red-team the systems within scope and report findings through a disclosure process.
• Help develop interpretability methods for frontier models and test their limits.

Skills

Python programming

Security research

Adversarial ML

Red-teaming

Technical writing

Attack surface analysis

Education

PhD in security or ML

Equivalent research experience

Job description

CommonAI CIC in Cambridge seeks an AI Security Researcher with an attacker mindset to test and secure AI systems used in regulated environments such as financial services. You will track relevant research, assess decisions from an attacker’s view, and help develop interpretable models and robust defenses.

You will threat-model AI systems, conduct red-team testing within scope, and share findings through clear briefings.