AI Safety & Assurance Lead

Location: London, United Kingdom; hybrid

Salary: £100,000–£120,000 per annum plus benefits

Company

Specialist UK technology business delivering advanced AI solutions into Defence, Government, and other highly regulated environments.

What you will do

• Lead technical AI safety, model evaluation, and assurance across a portfolio of AI products, with approximately 75% of the role focused on hands-on technical work.
• Design model evaluation frameworks and test harnesses; evaluate and benchmark LLMs against real-world use cases.
• Measure precision, recall, grounding, calibration, and hallucination, and stress-test models with adversarial and out-of-distribution data.
• Red-team systems for prompt injection, jailbreaks, data leakage, and tool misuse; test agent guardrails, permissions, and containment.
• Build automated evaluation into CI/CD pipelines, monitor deployed models for drift and regression, and support fine-tuning and model adaptation.
• Produce technical assurance evidence, present findings to senior stakeholders and customers, and contribute to audit, certification, and customer assurance activity.

Requirements

• Strong Machine Learning Engineering or Data Science background with hands-on experience taking AI systems into production.
• Production Python experience.
• Experience with PyTorch, Hugging Face Transformers, or similar technologies.
• Strong hands-on LLM experience, including RAG, vector search, prompting, tool use, and agent orchestration.
• Experience designing and running model evaluations.
• Experience with cloud deployment, containers, and CI/CD.
• Understanding of AI attack surfaces and failure modes.
• Applied statistics and experimental design knowledge.
• Current UK Security Clearance at SC level or eligibility to obtain SC.

Nice to have

• Experience in AI safety, adversarial machine learning, AI red teaming, technical assurance, or Defence/Secure Government environments.

Culture & Benefits

• Hybrid work arrangement based in London.
• Permanent position with benefits.
• Technical work on AI systems where reliability, safety, and trust are central concerns.