Senior Information Security Engineer with 15+ years securing federal platforms. When I'm not designing cloud security controls, I'm in my homelab breaking things on purpose to understand how they work. Here, I share what works, what fails, and the lessons I've learned since 2005.
"A robot may not harm humanity, or, by inaction, allow humanity to come to harm."
That nerdy kid in 1995 devouring Foundation and Empire under the covers with a flashlight? Yeah, that was me. Now I'm helping secure FedRAMP platforms while running local LLMs on my RTX 3090 to explore AI security. I'm implementing the actual controls to keep AI systems safe — thinking about the same questions Asimov posed, just with more YAML files, threat models, and security tooling than I expected.
Areas of expertise and interests
Implementing Zero-Trust architectures for federal systems. Microsegmentation isn't just theory – it's what stops lateral movement when something gets through.
Started with a $50 Raspberry Pi in 2015, now running enterprise gear in my basement. Every failure in my homelab is a lesson I don't have to learn the hard way anywhere else.
Running local LLMs for security research and AI safety experiments. That childhood obsession with Asimov's Three Laws evolved into real work on AI alignment.
73+ posts on security, AI safety, and homelab experiments — practical lessons from real projects, not just theory.
Security insights, AI experiments, homelab adventures, and lessons from the field.
The engineering story behind nexus-agents, a research-backed multi-model orchestration system that coordinates Claude, Gemini, and Codex through consensus voting, adaptive routing, and graph workflows.
How multi-model consensus voting catches blind spots that single models miss. The research behind adversarial roles, Bayesian aggregation, and structured deliberation across Claude, Gemini, and Codex.
How I went from naive round-robin model selection to a five-stage routing pipeline backed by RouteLLM, TOPSIS, and LinUCB research. The failures that led to each improvement.
Tools I'm building and maintaining.
A Model Context Protocol server that gives Claude direct control over Strudel.cc for AI-assisted music generation and live coding.
One tool to coordinate Claude, Gemini, Codex, and OpenCode. Routes tasks to specialized experts that collaborate on complex problems.
Battle-tested software development standards from real production systems. Start any project right in 30 seconds.
Got a security question? Building your first homelab? Debating if that CVSS 7.2 vulnerability is worth an emergency patch at 10 PM? Let's chat.