Profile
Adel ElZemity is a Postdoctoral Research Assistant in the SAYED Systems Group, School of Electronic Engineering and Computer Science, working on decentralised and federated AI for cyber security. He is completing his PhD in Computer Science at the University of Kent, where his research focuses on safe and deployable large language models for cyber security: the safety–performance trade-offs of fine-tuning LLMs on security data, small and efficient language models that can match or outperform large ones when orchestrated or distilled for edge deployment, and autonomous LLM agents for vulnerability management in industrial networks. His work has appeared at venues including RAID, ACM AISec, SECAI, DCOSS-IoT and IEEE iThings, and received the Best Student Paper Award at AI-SS 2026.
Teaching
Experience teaching undergraduate and postgraduate computer science, and supervising undergraduate and Master's projects, at the University of Kent. Associate Fellow of Advance HE (AFHEA).
Research
Research Interests:
AI security; LLM safety; decentralised and federated learning; intrusion detection; autonomous AI agents for cyber defence.
Publications
See Google Scholar / adelsamir.com.
Public Engagement
Research contributor to the EPSRC-funded CHARIOT project (Countering Harms Caused by Ransomware on the Internet of Things). Reviewer for the Journal of Computer Security.