🦉AI Security Architect · Vulnerability Intelligence
0000-0003-0901-3580About Me
I work where AI systems meet adversaries: securing the models, agents and pipelines organisations now depend on, and using AI to make cyber defence faster and more anticipatory. My background spans enterprise cyber defence, large-scale data engineering and a decade of academic research in machine learning and multi-objective optimisation.
In industry I have authored an organisation-wide AI security plan, built AI asset visibility and shadow-AI detection, red-teamed LLM guardrails before adoption, and run reliability and security for a large-scale vulnerability intelligence service. I hold a PhD in Intelligent Big Data Management, an MSc in Information and Communication Systems Security and a BSc in Mathematics from the University of the Aegean, with 7 publications and 177 citations.
- Security of AI & LLM systems (prompt injection, guardrails, agentic AI)
- Vulnerability & threat intelligence
- AI for cyber defence
- Machine learning & multi-objective optimisation
PhD in Intelligent Big Data Management
University of the Aegean
MSc in Information and Communication Systems Security
University of the Aegean
BSc in Mathematics
University of the Aegean
I work on the security of AI systems and on AI for cyber defence: AI security plans and controls, shadow-AI detection, prompt-injection testing and guardrail evaluation, and agentic AI threat modelling.
Alongside this I work on vulnerability intelligence, and my academic background is in machine learning and multi-objective (skyline / Pareto) optimisation over big data.
Please reach out to collaborate 😃
| Area | What I work on |
|---|---|
| Security of AI | AI security strategy and control design · prompt-injection and guardrail evaluation · agentic AI threat modelling · AI asset visibility and unsanctioned (shadow) AI detection |
| AI for security | Agentic workflows for security operations · ML detection engineering · AI-assisted code review and penetration-test reporting |
| Vulnerability & threat intelligence | Exploitation forecasting (KEV / EPSS) · advisory standards (CSAF) · CTI fusion with geopolitical and narrative signals |
| Decision intelligence | Multi-objective (Pareto / skyline) methods for ranking risk, choosing controls and briefing leadership under uncertainty |
