# Getting Started

> For AI agents: a documentation index is available at [/llms.txt](/llms.txt). Append `.md` to any page URL for markdown, or send `Accept: text/markdown`.

## Step 1: Install the SDK in your server app

Install the Statsig Azure AI SDK using the package manager for your language or framework.

#### NodeJS

```shell
npm i @statsig/azure-ai
```

#### Python

```shell
pip install azureai-statsig
```

#### .NET

```shell
dotnet add package StatsigAzureAI
```

## Step 2: Create a model deployment in Azure AI Studio

Log into your Azure AI Studio console and create a new deployment.

![Azure AI Studio model deployment creation interface](https://docs.statsig.com/images/integrations/azureai/getting-started/3172d6f9-176e-4a3b-817f-3a5bd3af4291.png)

After confirmation, Statsig takes you to the details page of that model.

![Azure AI model deployment details page](https://docs.statsig.com/images/integrations/azureai/getting-started/b8f46ed9-ce02-486f-b6f7-1d09f7b62f33.png)

Copy the **Target URI** (your _endpoint_ in code) and the **Key**. You need both in the SDK to call the APIs.

## Step 3: Get Statsig server SDK key

> **Info:**
>
> This guide assumes you have an existing Statsig account. Go to https://statsig.com/signup to create a new free account if you don't already have one.

Go to your **Project Settings** and choose the **Keys & Environments** tab. Scroll down to the **API Keys** section and copy the Server Secret Key. If one doesn't exist, create one or ask your project admin to create one.

![Statsig project settings API keys section](https://docs.statsig.com/images/integrations/azureai/getting-started/f60f6f2b-60d0-4804-8db5-b82d4ac96bcf.png)

## Step 4: Initialize Azure AI server

#### NodeJS

```js
import { AzureAI } from "@statsig/azure-ai";

await AzureAI.initialize("<STATSIG_SERVER_KEY>");
```

#### Python

```python
AzureAI.initialize("<STATSIG_SERVER_KEY>")
```

#### .NET

```csharp
using Statsig;
using Statsig.AzureAI;

await Server.Initialize("<STATSIG_SERVER_KEY>");
```