Infrastructure, Automation & Observability

I build platforms that run themselves, and prove it with data.

I design, automate and operate self-hosted infrastructure end to end: containers deployed from Git, network and DNS as code, single sign-on in front of every admin tool, and monitoring that tells me about a problem before the people using it do.

See the projectsAbout me

  • 22 projects
  • 49 containers in production
  • 50 automated health checks

Live from the lab

These projects run in production for friends and family. Current numbers from the public status feed:

Players online
…
Game servers up
…
Game platforms
…

Loading…

Full status page ↗

Featured case studies

The problem, the architecture, the decisions, and how it turned out.

Homelab Platform (GitOps)

49 containers in 12 Compose stacks on a self-hosted Unraid server, deployed from Git with secrets kept out of the repo, health checks on every service and weekly off-box backups.

  • Docker
  • Compose
  • Unraid
  • Git
  • GitHub Actions
Case study

Multi-Protocol Player Telemetry

A dependency-free Python collector that asks 12 game servers how many people are playing, over four binary protocols and four JSON APIs. It publishes counts-only JSON for a public page, richer detail for internal dashboards and Prometheus metrics for 90 days of history.

  • Python
  • Sockets
  • Prometheus
  • VictoriaMetrics
Case studyLive ↗

Corellia: Phantasy Star Online Server Platform

A three-instance private server platform (production, canary and legacy) for Phantasy Star Online: Blue Burst, migrated from Tethealla to newserv with staged rollouts, account and character migration between server generations, and imports from another community server.

  • C++
  • Python
  • Django
  • Docker
  • MySQL
  • GitHub Actions
Case study

What I work on