Skip to content
Hobbyist

What Hobbyist is

v0-alpha, not production ready

Hobbyist is a convenience layer over primitives you already have. Postgres, containers, and a reverse proxy are all free and mature. What managed platforms sell on top of them is instant provisioning, a connection string that just works, a dashboard that makes a database legible, and a deploy that happens on push. That convenience is the only thing missing from the self-hosted world, and it is the whole of what this builds.

One command gives you a project with a Postgres in it. Everything sleeps when nothing is using it and wakes when something connects. It runs on one box, with no Kubernetes, and you can walk away from it at any time with a single command.

Read this part before you install anything

Section titled “Read this part before you install anything”

This is v0-alpha. Postgres is the part that is reliably configured and working. The rest ranges from verified-but-young to openly broken, and the status page names which is which. There is no warranty and no support obligation. Do not put data you cannot afford to lose in here without your own backups, which today means pg_dump, because snapshots are built and not yet reachable.

The wedge

Section titled “The wedge”

Everything sleeps, and everything wakes on demand. This is the single reason for the project to exist, and it is the one thing no self-hostable alternative does. Self-hosted Supabase never sleeps. Xata’s open-source scale-to-zero plugin cannot wake a database, which is why reactivation stayed in their paid cloud. Coolify and Dokploy deploy apps well and never sleep them.

Sleep is what makes ten projects fit on one small box. Wake is what makes sleep invisible. How it works.

The three promises, in priority order

Section titled “The three promises, in priority order”
  1. You can always leave. The data directory is a plain Postgres data directory. pg_dump always works. hobby eject hands you a docker-compose.yml and the data, and gets out of the way.
  2. It runs on one box. A five dollar VPS, a Mac Mini, an old ThinkPad under a desk. If a feature requires a cluster, it is out of scope.
  3. It feels good. The reason people pay for managed platforms is ergonomics. Matching the primitives is not enough; matching the feel is the entire job.

When those conflict, the earlier one wins.

What it is not

Section titled “What it is not”

It is not a business. There is no cloud offering, no hosted tier, no paid feature, no metering and no billing, and no roadmap toward one. That is load-bearing rather than a disclaimer: it removes multi-tenancy, isolation hardening, usage accounting and quota enforcement from scope, and those removals are what make the project buildable by one person.

It also means several things are permanently out of scope, and adding them needs a decision record arguing the case: Kubernetes and clustering, multi-tenancy across different owners, end-user auth as a service, realtime subscriptions, edge execution, DNS and secrets management.

Where to go next

Section titled “Where to go next”