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

# Connect your IDE to a cluster

> Use VS Code, Cursor, JetBrains, or Jupyter to edit and run code on remote GPU hardware.

Use VS Code, Cursor, JetBrains, or Jupyter to edit and run code on a remote Acasia cluster.

**In this guide:**

* Every IDE connection — VS Code, Cursor, JetBrains, Jupyter — runs over the same SSH foundation.
* If `ssh aix@<host>` doesn't work, fix that before configuring an IDE.
* Use an `~/.ssh/config` alias and let your IDE point at it for repeatable, key-aware connections.

## When to connect to a cluster

| Use case                        | Description                                                              |
| ------------------------------- | ------------------------------------------------------------------------ |
| Inspect the runtime environment | Check GPU status, installed packages, files, logs, or running processes. |
| Run commands manually           | Execute shell commands directly on the remote cluster.                   |
| Develop remotely                | Use a local IDE while code and commands run on the Acasia cluster.       |
| Debug workloads                 | Investigate model-serving issues, dependency errors, or resource usage.  |
| Run notebooks                   | Start Jupyter on the cluster and access it from your browser.            |
| Validate GPU access             | Confirm that the cluster can see assigned GPUs with `nvidia-smi`.        |

## Connection options

| Method            | Best for                                                                      |
| ----------------- | ----------------------------------------------------------------------------- |
| SSH terminal      | Fastest way to verify access, inspect the machine, or run commands.           |
| VS Code           | Full remote development with editor, terminal, extensions, and file explorer. |
| Cursor            | AI-assisted remote development on the cluster.                                |
| JetBrains Gateway | IntelliJ, PyCharm, WebStorm, and other JetBrains remote workflows.            |
| Jupyter           | Notebook-based development, data science, and experimentation.                |

## Before you connect

| Requirement               | Description                                              |
| ------------------------- | -------------------------------------------------------- |
| Correct organization      | You are operating in the intended Acasia organization.   |
| Active cluster            | The target cluster status is Active.                     |
| SSH key attached          | Your public SSH key is attached to the cluster.          |
| Private key available     | The matching private key exists on your local machine.   |
| Connection command copied | You copied the SSH command from the cluster detail page. |
| Network access            | Your local network can reach the cluster host and port.  |

## Find the cluster connection command

1. Open Developer Portal → **Clusters**
2. Find the target cluster and open the actions menu
3. Select **View Details**
4. Locate the **Connection** panel and copy the SSH command

```bash theme={null}
ssh aix@203.0.113.42
```

## Recommended first test

Before configuring an IDE, verify the plain SSH connection works:

```bash theme={null}
ssh aix@203.0.113.42
hostname
nvidia-smi
```

<Warning>
  If plain SSH does not work, fix that first. IDE connections rely on the same SSH foundation.
</Warning>

## Security practices

<Warning>
  Only upload public SSH keys to Acasia. Never paste private SSH keys into Acasia, email, chat, tickets, or shared documents.
</Warning>

* Remove SSH keys that are no longer needed
* Use separate SSH keys for personal devices, automation systems, and temporary support access
* Review cluster access after team changes or device replacement

## Connection guides

<CardGroup cols={2}>
  <Card title="Connect with SSH" icon="terminal" href="/get-started/connect-cluster-ssh">
    Direct terminal access — the base for every other method.
  </Card>

  <Card title="VS Code Remote SSH" icon="code" href="/get-started/connect-ide-to-cluster">
    Full editor, terminal, and extensions against the cluster.
  </Card>

  <Card title="Cursor" icon="arrow-pointer" href="/get-started/connect-ide-to-cluster">
    AI-assisted development against the cluster.
  </Card>

  <Card title="JetBrains Gateway" icon="j" href="/get-started/connect-ide-to-cluster">
    PyCharm, IntelliJ, WebStorm remote backends.
  </Card>
</CardGroup>