Research
My research sits at the intersection of artificial intelligence and cyber security: how AI systems can make security operations faster and more reliable, and how we measure whether they actually do.
Publications
AI-Powered System for an Efficient and Effective Cyber Incidents Detection and Response in Cloud Environments
IEEE Transactions on Machine Learning in Communications and Networking
An AI-driven approach to detecting and responding to security incidents in cloud environments, aimed at reducing the manual effort involved in triaging and acting on cloud-native telemetry.
Read the paper →Research Interests
- AI for Cyber Security Applying language models and agentic systems to defensive security work.
- Security Agents Multi-agent architectures that investigate and reason over security data.
- LLM Evaluation Benchmarking how well frontier models actually perform on security tasks.
- Incident Response Hypothesis-driven triage, enrichment, and case management.
- Digital Forensics Automating evidence collection and analysis.
- Reinforcement Learning Training and evaluating defensive agents in simulated networks.
Research in Practice
Projects that put these interests into working code.
- Cyber LLM Benchmark Hub Tracking frontier model performance across cyber security benchmarks.
- Console IR Terminal-first incident investigation over standardised OCSF events.
- tie-hunt Predicting an adversary's unobserved MITRE ATT&CK techniques, with agents validating each lead.
- Cyber Defense Sim A Gymnasium cyber range for training defensive RL agents.