What we're
actually building.
Short, honest writing on the projects we're shipping and the tradeoffs behind them: reverse engineering, automation, infrastructure, AI, and the occasional game devlog.
Where a DDoS defense sits decides what it can save
We got floods just from developing game servers, so we built a lab to test the obvious idea: when you're attacked, only serve clients you already know. It works, and then it doesn't, and the reason why is the whole point.
One button, one invoice: wiring a CRM to Stripe
The gap between 'we agreed on a price' and 'the customer has an invoice' is usually a human copy-pasting between two systems. We closed it with a small bridge.
Running a company on hardware we own
Instead of a dozen SaaS subscriptions, our whole internal stack runs on Proxmox boxes we own, behind one login, with backups that have actually been restored.
Devlog: descending into Aethelgard
Our flagship game is an action-RPG built around a permadeath descent. Here's where the prototype actually is, and the discipline keeping it from swallowing the studio.
Start with the boring version
Most useful software starts as a boring first version: a form, a table, a script, a dashboard, or a workflow that saves somebody time.
AI is a tool, not a plan
AI is useful when it reduces repetitive work or catches mistakes. If a normal script is cleaner, use the normal script.
Self-hosting is a tradeoff
Self-hosting can be cheaper and more controllable, but it only works if backups, updates, monitoring, and recovery are treated as part of the project.