Director of Frontier AI Security (Contract)

About the role

We are partnering with a global financial services organisation to deliver a high‑profile Frontier AI security project. The contractor will join a senior technical team in London on a hybrid basis, working three days per week on‑site for an initial six‑month period.

Key responsibilities

• Design and implement deep security engineering solutions for frontier large language models (LLMs).
• Conduct evaluations, agentic loop testing and failure‑mode analysis of cutting‑edge AI systems.
• Develop and execute exploit and offensive‑security assessments to harden AI‑driven products.
• Collaborate with product and leadership teams to ship production‑ready AI security features.
• Provide expert guidance to strengthen the organisation’s overall cybersecurity posture.

Required profile

• Very senior hands‑on technical specialist with a track record of delivering production AI systems.
• Strong background in AI, cybersecurity and software engineering.
• Experience with frontier AI models and large language models beyond traditional machine learning.
• Proven expertise in exploit development and offensive security.
• Ability to operate at senior leadership level and influence security strategy.

Required skills

• Artificial Intelligence (AI)
• Cybersecurity
• Software Engineering
• Large Language Models (LLMs)
• Frontier AI
• Exploit Development
• Offensive Security
• Security Engineering

What we offer

• Competitive daily rate of £900‑£1000 inside IR35 via umbrella.
• Hybrid work arrangement with three days per week on‑site in London.
• Opportunity to shape security for cutting‑edge AI in a leading financial services firm.