Giving agents access to realistic data via branches should scale the feature development without increasing costs.

This shows how you can use the “scale to zero” feature of the Xata platform to improve feature development while keeping costs in check.

AI agents are “clever” to use available tools like grep, sed to find the relevant code files based on a prompt.

Here we experiment with giving AI agent tools to change environment variables in a repository that it checked out and it worked.

We explored giving AI agents access to compute (via Sandbox) and database via Xata branches and some “commandments” to follow a developer-like workflow.

Thinking about changing your data model?
View your schema as a table or UML diagram.
When you are ready, use xata roll to apply changes without downtime.

Shipping Text to SQL you can trust isn’t just prompting.
It’s engineering for simplicity.
At Xata we use a tiny context, pass schema + current SQL, a focused system prompt, and the right model per task.
Fewer moving parts, cleaner UX.
Build notes → xata.io/blog/how-we-...

Data exploration is easier with the ability to generate SQL from a simple question.
No need to worry about syntax.
Just dive into the data.

Branch metrics provide insight into CPU, memory, throughput, egress, and more.
Use interactive charts to view trends over time or investigate anomalous spikes.
Metrics are stored for up to 60 days at no additional cost.

Have a branch you don’t need running all the time?
Save money by setting scale-to-zero as the default for a project or override it in the branch settings.

Our list of supported extensions is growing every day.
Got any requests?

You have full control over your branches directly from the console.
Adjust the instance size, add replicas, tune compute settings, and modify your Postgres config.

In Xata, every branch is a full Postgres instance.
Use it as a staging replica with PII automatically obfuscated, a production-ready database, or an ephemeral copy for a quick PR.