Blog ·
The nine pieces of infrastructure behind an AI-built app that you never set up
Deploying an app used to mean assembling a stack. White Ghost lists the nine pieces it manages for every app, each with live status, in the Overview tab.
Ask an engineer what it takes to put a web app in production and you get a list: a build pipeline, a database, a repository, hosting, credentials. Ask someone who just built an app with an AI assistant and the honest answer is: no idea, the assistant wrote the app, not the infrastructure.
White Ghost makes that list explicit. The Overview tab of every app has a section called "What makes up this app": the nine pieces the platform provisions and runs for you, each with its status and last update.
The nine pieces
- Deployment pipeline. Turns every push to the repository into a build and a rollout. Real builds of a full app measure around a minute.
- Database. A managed PostgreSQL database, provisioned with the app.
- Build accelerator. Keeps those builds fast between deploys.
- App registry entry. The app's record in the platform: what it is, where it runs.
- Preview database. The development-side database that local development and pull request previews run against, kept separate from production.
- Code repository. A normal repository in your own GitHub account. Also downloadable from the console as a .zip at any time.
- Starter code. The scaffold the app began from, including the instructions that teach AI assistants its conventions.
- Live service. The running app itself, on its URL, scaling down to zero when idle and back up with traffic.
- App identity. Who this app is inside the platform, which is what lets integrations inject credentials securely without the code ever holding a key.
Why show the plumbing at all
Because invisible infrastructure is only comforting until something needs attention. Each piece carries its own status, so "is everything okay?" has a concrete answer instead of a shrug. And because seeing the list changes how builders think: an app is not a chat artifact, it is a small production system, and this is what one actually contains.
The inverse reading is the point: this is the checklist you did not have to assemble. Anyone who has wired a pipeline, a database, secrets, and hosting by hand knows each line hides an afternoon. On White Ghost the assistant writes the app, and these nine show up already running.
Frequently asked questions
Where can I see these pieces?
In the Overview tab of every app in the White Ghost console, under "What makes up this app". Each piece shows its status and when it was last updated.
Do I ever have to configure them?
No. They are provisioned when the app is created or when the app first needs them, and the platform operates them. Your work happens in the code and in the console's app-level controls: secrets, domains, access.
Is the code one of the managed pieces, and is it locked in?
The code repository is managed in the sense that the platform creates and wires it, but it lives in your own GitHub account, and the console offers a .zip download of the source at any time. You can leave with the code and the history.