One backend platform to build, run and govern AI-built software

Your coding agent writes your backend in TypeScript with the Xano SDK. Run it on the Xano Engine on your machine, then ship it to Xano through guardrails laid down ahead of every change.

Start free Explore the SDK $ npx @xano/sdk init my-app
  • Lead car: Xano Engine, projecting its own track ahead, with an AI coding agent conducting from the cab
  • The agents (Claude Code, Codex, Cursor) take turns: each writes an SDK call, and a car couples on for that primitive: Tables, APIs, Tasks, MCP servers, Agents, API groups, Functions, Middleware, Triggers, Tools, MCP prompts, Knowledge, MCP resources, Realtime, Workflow tests, Addons, Microservices
  • A second engine branches off to build in parallel, then rejoins

What is Xano?

Xano is one backend platform to build, run and govern AI-built software. Your coding agent (Claude Code, Codex or Cursor) writes the database, APIs, MCP servers and agents in TypeScript with the Xano SDK. You run them on the Xano Engine on your machine, then deploy them to Xano, where one set of rules governs every release.

In production

Business-critical systems in production, not prototypes.

Deriv

3M+ users across 150 countries

  • 100 transactions/sec
  • 100K daily active users

Heimstaden

€80B in investment properties

  • 32K rental units managed
  • Built for a highly regulated market

Generali

$1M+ saved annually

  • 30+ countries
Running on Xano
  • Fluence
  • BNP
  • Deriv
  • Aligned
  • Siemens
  • MRI
  • Heimstaden
  • Concentrix
  • Europ Assistance
  • Choice Digital
  • Fanova
  • College Hunks
The route · from an empty folder to production, on one line

From npx @xano/sdk init
to production in three commands

Write with AI · Run locally · Deploy to Xano. The code you ran on your own machine is the exact code that goes live.

01

Write with AI

Scaffold the project, then your coding agent writes the backend and frontend in TypeScript with the Xano SDK.

npx @xano/sdk init my-app
02

Run locally

Your backend runs on the Xano Engine on your machine. No account, no network.

npm run xano:deploy
03

Deploy to Xano

The same code goes live on Xano, through the same guardrails.

xanosdk deploy --ephemeral
On rails means guardrails

One set of rules for every person and every agent.

Agents write a lot of code, fast. Xano lays the checks down ahead of every change, so speed never means going off the rails.

  • Types end to end

    Rename a column and every consumer breaks at compile time, not in production.

  • Strict export in CI

    Fail the build on any warning before a change can reach a workspace.

  • Stable identities

    xano.lock keeps renames as renames, so your public URLs stay yours.

  • Preflight round-trip

    Import into a fresh environment, export it back and diff before you ship.

  • Release gates

    Promotion reads the release back and refuses anything that did not arrive.

  • One-command rollback

    Back up the branch on release and set the old one live if you need to.

Anatomy of the train

Database, APIs, MCP servers and agents, on one platform.

Everything in one place, with nothing to assemble. The engine runs it, the track guards it, and the cars carry what you build.

01 Lead car

Xano Engine

The runtime that pulls everything. A single binary with Postgres built in that runs on your machine, in CI, or in Xano’s cloud.

How the Xano Engine runs locally
02 Track

Xano SDK guardrails

Types end to end, strict exports, stable identities and release gates. The track appears exactly where you are going.

SDK guardrails
03 Cars

What you build

APIs, realtime channels, background tasks, MCP servers, AI agents and tables. Couple on as many as you need.

Platform overview
04 Conductor

Autopilot on Kubernetes

Keeps production on schedule: right-sizes resources, scales and recovers so the train never stalls.

Autopilot
05 Signal box

Xano Studio

Directs your coding agents, reviews every change and only gives the green light when tests and review pass.

Xano Engine · on your laptop

Run your backend locally on the Xano Engine

The whole Xano backend, packaged to run on your laptop. It starts in seconds, offline, and is pinned like a dependency, so every checkout runs the same engine.

Footprint
Single binary, Postgres built in
Runs on
macOS (Apple silicon) and Linux
Needs
No account, no network
Ships to
Xano cloud with one command
Questions

Xano questions, answered

What kind of tool is Xano?

Xano is a backend platform: a managed PostgreSQL database, REST APIs, realtime channels, background tasks, MCP servers and AI agents, plus the rules, releases and request history to run them in production. It is frontend-agnostic, so React, Next.js, mobile apps or any front end can call its APIs, and it runs in Xano’s cloud or yours.

Can I run Xano locally?

Yes. The Xano SDK runs your whole backend on the Xano Engine on your machine: a single binary with Postgres built in, for macOS (Apple silicon) and Linux, with no account and no network. Run npx @xano/sdk init my-app, then npm run xano:deploy. One command deploys the same code to Xano.

Does Xano support TypeScript?

Yes. With the Xano SDK (@xano/sdk on npm, MIT licensed), tables, APIs, realtime channels, tasks, MCP servers, agents, middleware and triggers are TypeScript objects in your repo. A generated route manifest types every call end to end, so a backend rename is a compile error, not a 404. The agent skill teaches Claude Code, Codex and Cursor to write it.

How is Xano different from Supabase?

Supabase is an open-source developer backend (Postgres, auth, storage, edge functions) where review lives in your own pipeline. Xano is one platform to build, run and govern AI-built software: Postgres plus one set of workspace rules enforced during the build, immutable releases promoted through review gates, and managed, BYOC or self-hosted operations. Read the Xano vs Supabase comparison.

Is Xano free?

Yes. The Free plan costs $0 a month and includes 1 workspace, a Postgres database with up to 100K records, and unlimited REST APIs (rate limited). Running your backend locally on the Xano Engine needs no account at all. Paid plans start at $85 a month, billed yearly. See Xano pricing and plans.

AI builds software. Xano makes it trustworthy.

Start free. Add dedicated infrastructure when production needs it.