I’m Michael Osburn — security architect, and maritime history buff.

This blog documents the ongoing project of turning a home network into a real engineering environment: reproducible, auditable, self-contained, and useful for the kind of work I actually care about — AI tooling and security research. Overtime I am sure this will end up blending my history research with my tech research.

The infrastructure is managed entirely through Ansible, version-controlled, and built with the same discipline I’d apply to production systems. Every role has tests. Every change goes through version control. The lab can be torn down and rebuilt from scratch.

The goal isn’t complexity for its own sake. It’s the opposite: eliminating the kind of “I set that up two years ago and have no idea how” drift that makes home networks brittle and home labs useless when you actually need them.

What runs here

  • Keycloak — single sign-on for every service in the lab
  • Forgejo — self-hosted git, no subscription, no rate limits, no telemetry
  • Redmine — project and issue tracking for lab work
  • Docmost — internal documentation, also SSO-integrated
  • Ollama + Claude CLI — local and cloud AI, both available from the same workstation
  • TheHive + Cortex — security incident and threat intelligence tooling
  • ELK stack — centralized logging across all hosts

The rules

  1. If it’s not in Ansible, it doesn’t exist.
  2. If it’s in Ansible, it has a test.
  3. Secrets go in a vault, not in a post.
  4. AI is never to be used with history, only tech.