# Launch a dev environment (/docs/guides/developer-workspace/dev-environment/launch-a-dev-environment)



A dev environment is a running server with your chosen IDE, compute resources, and storage volumes attached. Each dev environment runs independently and persists your home directory between sessions.

Before you start [#before-you-start]

You need access to Developer Workspace. If you cannot see **Notebooks** in the left sidebar, contact your workspace administrator.

Create a dev environment [#create-a-dev-environment]

1. Select **Notebooks** in the left sidebar.

2. Select <kbd>+ New Notebook</kbd> in the top-right corner.

3. Enter a **Name** for your dev environment. Use lowercase letters, numbers, and hyphens only.

4. Choose an IDE from the **IDE type** dropdown. See [Choose an IDE](choose-an-ide) for guidance.

5. Select a base image from the **Custom Notebook** dropdown (for example, `jupyterlab:python3.12`).

6. Set **Minimum CPU** and **Minimum Memory Gi** to match your workload. See [Configure a dev environment](configure-a-dev-environment) for sizing guidance.

7. Expand <kbd>Advanced Options</kbd> if you need GPU access, and set the number of GPUs and vendor.

   <Callout type="info" title="GPU hardware required">
     GPUs are only available if your BullSequana AI installation includes GPU-enabled hardware. Contact your workspace administrator if this option is missing.
   </Callout>

8. Review the **Workspace Volume** section. A 5 Gi home directory volume is created automatically. Adjust the size if needed.

9. Select <kbd>LAUNCH</kbd>.

Wait for the dev environment to start [#wait-for-the-dev-environment-to-start]

After launching, your dev environment appears in the list with a spinning indicator. Starting typically takes one to two minutes.

When the status changes to a green check mark, the dev environment is ready.

Connect to your dev environment [#connect-to-your-dev-environment]

Select <kbd>CONNECT</kbd> next to your dev environment. Your IDE opens in a new browser tab.

Your home directory is mounted automatically. Any files you save there persist between sessions.