@Doxas
About
Teen self-taught AI security researcher. I build autonomous red-teaming agents, find vulnerabilities in AI systems, and think about why current security frameworks are fundamentally incomplete. Writing at the intersection of offensive security, complexity theory, and machine reasoning.
Available for
Freelance security auditing and AI Red Teaming
Doxas's blogs
Recently published
Attack Graph Simulator: Reducing Uncertainty in Security Through Bayesian Inference
Jan 25 · 5 min read · Security systems are complex. Attackers don’t follow predictable paths and defenders rarely have complete visibility into a system’s state. This uncertainty makes reasoning about attacks difficult — e
Join discussion
When Reinforcement Learning Is The Wrong Tool
Jan 5 · 3 min read · I recently worked on a small game that looked deceptively simple but revealed an important lesson about how we choose tools in AI. The task was this: You must guess a 4-digit code with no repeated digits.After each guess, you receive feedback: Dead:...
Join discussion
Why Critical Thinking Is Becoming the Developer’s Most Valuable Skill 🧠⚙️
Dec 6, 2025 · 4 min read · Modern development is no longer just about writing functions and fixing bugs.It’s about reasoning under uncertainty, understanding systems, questioning defaults, and seeing what others overlook. Your environment, your tools, the libraries you trust, ...
Join discussion