AI Security Engineer

About the Role

We are looking for someone who understands how modern AI systems work — and is eager to do something useful with that knowledge.

This role sits across two areas: helping our teams adopt and apply AI in their day-to-day work, and contributing to how we keep those AI systems secure. You will work directly with both technical and non-technical teams, translating business processes into practical AI-powered solutions, while also supporting our security posture around AI adoption.

You will have real ownership from day one, with experienced colleagues to work alongside. This space is new enough that curiosity and drive matter as much as a polished CV.

What You'll Do

AI Enablement & Automation

• Work with teams across the organisation — including non-technical ones — to identify where AI can meaningfully improve how they work.
• Design and build AI-powered automations and workflows, turning business requirements into working solutions.
• Embed directly with teams to understand their processes, then prototype and iterate quickly.
• Help teams understand what AI can and cannot do, and guide them in using it effectively and responsibly.
• Document patterns, playbooks, and reusable components so that knowledge scales beyond individual projects.

AI Security

• Participate in security reviews of AI-enabled applications, agents, and LLM-integrated workflows.
• Help identify vulnerabilities in LLM-based systems: prompt injection, data leakage, insecure tool use, and authentication gaps.
• Support evaluation of agentic workflows for privilege escalation, unsafe action chaining, and unintended side effects.
• Assist with assessing how third-party AI platforms (e.g., Claude, OpenAI APIs, internal AI tools) are integrated and used across the organisation.
• Apply frameworks such as OWASP LLM Top 10 and MITRE ATLAS to help identify and prioritise risks.
• Help maintain an inventory of AI tools in use and support internal governance and policy documentation.
• Contribute to building test harnesses and repeatable validation frameworks.
• Work with engineering teams to translate findings into actionable mitigations.
• Stay current — this space moves fast, and you'll be expected to keep pace.

What We're Looking For

Must-Haves

Hands-on experience with LLMs, AI APIs, or AI-powered tooling — you have built or experimented with something using them.

Able to communicate clearly with non-technical people and translate what you are working on into plain language.

Can write code or scripts to get things done — language is less important than the ability to ship.

Exposure to LLM security concepts — prompt injection, jailbreaks, RAG risks — even from personal research or reading.

Curious, self-directed, and comfortable figuring things out without a detailed playbook.

Nice-to-Haves

Experience building automations or integrations using AI tools

Familiarity with agent orchestration, tool calling, or multi-agent systems.

Exposure to application security fundamentals: OWASP, secure SDLC, code review.

Any cloud platform exposure (AWS, Azure, GCP).

CTF participation, bug bounty, or independent security research.

Incident Response experience; lead or in a front-line IR team

Knowledge of AI governance or compliance frameworks.