Skip to main content
By the end of this guide you will have a running service deployed at https://<your-endpoint>.svc.uniac.ai.

​
Prerequisites

  • Python 3.10 or newer
  • Docker Desktop (for the local simulator)
  • A Uniac account — sign up at api.uniac.ai

​
1. Install the SDK and CLI

pip install uniac
curl -fsSL https://install.uniac.ai | sh
Verify the install. 
uniac --version

​
2. Log in

uniac auth login
This opens a browser window and completes the OAuth flow.

​
3. Create a new service

uniac init my-first-service
cd my-first-service
uniac init scaffolds a minimal Python service. Open the generated file and add a handler.
service.py
from uniac import FastAPIService

class Greeter(FastAPIService):
    def routes(self):
        @self.get("/hello")
        def hello(name: str = "world"):
            return {"message": f"Hello, {name}!"}

​
4. Run it locally

uniac sim
The simulator starts your service in a container. Call it from another terminal. 
curl http://localhost:8000/hello?name=there
# {"message":"Hello, there!"}

​
5. Deploy

uniac build
uniac deploy
Once the deploy completes, the CLI prints the public endpoint. 
Deployed: https://greeter-ab12cd.svc.uniac.ai
Hit it from anywhere. 
curl https://greeter-ab12cd.svc.uniac.ai/hello?name=cloud

​
Next steps

Your first service

A deeper walkthrough with explanations.

Service-to-service calls

Have one service call another.

Add AI primitives

Use Uniac’s LLM building blocks.

Custom domains

Map your own domain to a deployed service.