Large-language-model (LLM) agents can autonomously plan, tool-use and self-improve, creating fresh attack surfaces that fall between classical software bugs and adversarial ML.

The post-holder will help the project team discover, characterise and mitigate vulnerabilities in such agents and you will support the project team in exploring these questions. Typical duties include implementing proof-of-concept exploits and defences, collecting/curating data, and contributing to papers, demos and industry outreach.

The post is ideal for an MSc graduate (or exceptional BSc Hons) passionate about AI and Cybersecurity who wants to gain deep, publishable experience before a PhD or industrial R&D career.

About The Person

The successful candidate must have, and your application should clearly demonstrate that you meet the essential criteria and where appropriate meet the desirable criteria.

• A minimum 2:1 degree in Computer Science, Cyber Security, AI, or a related field
• Experience working in an academic research environment
• Significant Python programming experience
• Familiarity with at least one LLM framework or API (Hugging Face, llama.cpp).

To be successful at shortlisting stage, please ensure you clearly evidence in your application how you meet the essential and, where applicable, desirable criteria listed in the Candidate Information on our wesbite.

The post is available on a fixed term contract for 12 months. Fixed term contract posts are available for the stated period in the first instance but in particular circumstances may be renewed or made permanent subject to availability of funding.