beane.me

Patrick Beane · Platform, Infrastructure & Security Engineer

I build infrastructure that defends and repairs itself, and I run it in public.

Security, automation, and recovery for multi-node platforms: automated threat response, backups that are proven by restore tests, GitOps, and DNS failover. Everything on this page is running in production right now, and you can check it yourself.

HealthyLive from my production platform

Everything is operating normally.

7 of 7 services are healthy; 8 of 8 fleet nodes are reporting; 3 of 3 backups are on schedule. 1 item to watch.

Services healthy
7/7
all with automatic DNS failover
Bans yesterday
279
a typical day (median 278)
Backups on schedule
3/3
oldest run 5.8 h ago · 3 restore-tested
Uptime, 7 days
99.98%
12 of 12 monitors up
Open the full operations briefing

Featured work

Operations Briefing

The evidence behind the status above.

The panel at the top is the summary. The full briefing shows how each number is earned: which attack techniques were seen and how the decision engine escalated them, restore tests for every backup, standby readiness for failover, and GitOps and CI status. Every line is computed from the platform’s own APIs, not written by hand.

  • Python
  • FluxCD
  • CrowdSec
  • Renovate
  • Grafana
  • Gitea Actions
Open ops.beane.me

tfSTRIDE

See security risks before you deploy.

Upload a Terraform plan and get a STRIDE threat model back: findings with evidence, trust boundaries, and the controls already in place. Works across AWS, Azure, and GCP with no cloud credentials and no LLM, held in place by thousands of regression tests.

  • Python
  • Terraform
  • AWS
  • Azure
  • GCP
Open tfstride.beane.me

The platform

What the briefing is built on

Each piece below is its own public service. The seven counted above all fail over automatically by DNS; what fails over differs, and each card says how. Git is the exception: it runs a fenced manual standby, because automatic failover isn’t safe for it.

Writing

Selected case studies

How these systems were designed, what broke, and what I changed because of it.

All 16 case studies

About

Where this started

My introduction to Linux was a freshly mailed Ubuntu 6.06 CD when I was twelve. By my late teens I was securing my own VPSes, and I never really stopped: the platform above is the same curiosity, grown into something that runs in production.

I care about building systems where security, automation, and accessibility are part of the foundation rather than added later, and about explaining them clearly enough that the next person can run them. Outside the platform, I built Maggie.lol, a memorial for two dogs who meant a lot to me, and Hermes, a small rate-limited jokes and quotes API.

I also take on a limited amount of infrastructure consulting for small technical teams.