Launching today
instancez
The Backend for Developers and Agents
9 followers
The Backend for Developers and Agents
9 followers
instancez is a single container backend defined in one YAML file and is supabase-js-compatible. Declare tables, auth, row-level security, storage and functions, and it migrates your Postgres database for you. It ships as a single binary or container with a built-in dashboard, keeps your supabase-js code working, and promotes the same file from dev to staging to prod. The YAML is also easy for LLMs and coding agents to read and write and do local & cloud development and deployments!
Free Options
Launch Team / Built With
Customer.ioAutomate Messaging Everywhere — Startups Get 12 Months Free
Promoted
Hey Product Hunt, I'm Sa'd, one of the makers of instancez.
Every backend I built started the same way: set up auth, an API, storage and access rules, usually as a multi-container stack with a pile of migrations on top. Doing that once is fine. Doing it again for every branch and every experiment is what wore me down.
So instancez is the whole backend in one YAML file, run by one binary or container in front of your Postgres. There's no stack to bring up, so I can run several environments side by side, one per branch, and iterate on each of them in parallel. A new environment is the same file and a start command, not another round of bootstrapping.
That also makes it portable. The same file runs on a laptop, on your own servers, in Docker or on instancez Cloud. Your supabase-js code keeps working when you point it at the new URL.
It's friendly to coding agents, and to LLMs in general, mostly because YAML is easy for them to read and generate. instancez diffs the file against your database and applies only the change, so there are no migration files to write. The built-in dashboard writes edits back to the YAML as a diff you can review.
Install the skill with npx skills add instancez/instancez and your coding agent learns the YAML syntax, the RLS patterns and the inz CLI.
To try it: install from get.instancez.ai, then `inz init` followed by `inz dev --embedded-pg`. To host it yourself use `inz serve` or Docker, or deploy to our cloud with `inz cloud deploy`.
How do you handle parallel environments today, and what would you want your coding agent to build on this first?