> ## Documentation Index
> Fetch the complete documentation index at: https://docs.negentro.tech/llms.txt
> Use this file to discover all available pages before exploring further.

# Get started with persistent AI memory

> Give your AI application a reliable way to store information, retrieve relevant history, and assemble context across interactions. Start with the SDK or connect directly through the REST API.

## What you'll build

In this quickstart, you'll learn how to:

## Before you begin

You'll need a:

* Negentro API key
* A supported runtime
* An application that can make authenticated API requests.

## Steps

```bash terminal theme={null}
npm install @negentro/sdk
```

```bash terminal theme={null}
pip install negentro
```

```typescript client.ts theme={null}
import { NegentroClient } from "@negentro/sdk";

const client = new NegentroClient({
  apiKey: process.env.NEGENTRO_API_KEY!,
});
```

```python client.py theme={null}
from negentro import NegentroClient
import os

client = NegentroClient(
    api_key=os.environ["NEGENTRO_API_KEY"]
)
```

```typescript example.ts theme={null}
const memory = await client.memory.store({
  user_id: "user_123",
  content: "Prefers concise technical explanations.",
});
```

```typescript example.ts theme={null}
const context = await client.context.retrieve({
  user_id: "user_123",
  query: "How should I explain this technical topic?",
});
```

## What happens next?

Your first integration is the starting point. As your application grows, you can configure retrieval behavior, manage user and session scopes, connect data sources, and observe system performance.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.