AI8 min
Teaching Defensive AI to Think Like a Red Team
A walkthrough of how I architected multi-agent simulations that pit autonomous attackers against defensive RL policies to harden enterprise networks.
Read article →
Documenting the experiments and ideas behind resilient AI security
Deep dives on reinforcement learning, intent-aware defence, and the architectural decisions powering projects like pentestMCP and North-Star.
6 posts
6 minMCP
Designing pentestMCP for Autonomous Offensive Ops
Breaking down the architecture decisions that let pentestMCP compose reconnaissance, exploitation, and reporting without drowning operators in noise.
Read →
7 minIoT
Real-Time IoT Botnet Detection Using Graph Neural Networks
How graph neural networks can identify coordinated botnet behaviour across distributed IoT devices by modelling communication patterns as spatial-temporal graphs.
Read →
8 minAI
Explainable AI for Exoplanet Discovery: Lessons from NASA Space Apps
How we built North-Star, an interpretable ML pipeline that helps astronomers understand why a model flags a star system as hosting exoplanets.
Read →
9 minKubernetes
Zero-Trust Kubernetes: Hardening Cloud-Native Infrastructure
A practical guide to implementing zero-trust principles in Kubernetes clusters, from network policies to runtime threat detection.
Read →
6 minAI Security
Adversarial Attacks on Large Language Models: What Developers Need to Know
Exploring prompt injection, data poisoning, and model extraction attacks against LLMs—plus practical mitigations for production deployments.
Read →
End of feed — more soon