Configure your Databricks project

The Databricks IDE extension offers a Configuration view within the extension that allows you to easily configure and update settings for your Databricks project. These features include local folder settings, AI tools installation, a target workspace deployment selector, easy configuration of authentication and compute, workspace folder sync, and simple steps for activating the Python virtual environment necessary for debugging.

The Configuration view within the Databricks IDE extension is available once you have created or migrated a project to a Databricks project. See Create a new Databricks project.

Note

Earlier versions of the Databricks IDE extension defined configuration settings in a project JSON file and environment variables were set in the terminal. In the release version, project and environment configuration is found in the databricks.yml and databricks.env files.

If your project is a Databricks Asset Bundle, the Databricks extention UI also provides a Bundle Resource Explorer and a Bundle Variables View to manage your bundle resources and variables. See Bundle and explorer features.

Set the local folder

To select a different local folder for your Databricks project, click the gear icon associated with the Local Folder entry in the Configuration view.

Install and use AI tools

Important

This feature is in Beta.

The AI tools section of the Configuration view enables you to install AI tools so that your coding agents work effectively with Azure Databricks. The extension checks whether AI tools are installed at the global or project scope, including installations that you performed through the Azure Databricks CLI, and updates the displayed status. For more information about AI tools, see Agent skills for AI coding assistants and aitools command group.

To install AI tools:

1. In the Configuration view of the extension, click Install AI tools.
2. Follow the prompts to select a scope and tools to complete the installation.

When AI tools are installed, a green robot icon appears next to AI tools. AI tools automatically update to the latest version.

To uninstall AI tools, right-click AI tools and select Uninstall AI tools.

Change the target deployment workspace

To select or switch the deployment target for your Databricks project (for example, to switch from a dev target to a prod target):

1. In the Configuration view of the Databricks extension, click the gear icon (Select a Databricks Asset Bundle target) associated with Target.

   

2. In the Command Palette, select the desired deployment target.

Once a target is configured, the Host and deployment Mode is displayed. For information about Declarative Automation Bundles deployment modes, see Declarative Automation Bundles deployment modes.

The workspace host can be changed by modifying the target workspace setting in the databricks.yml configuration file associated with the project. See targets.

Note

The following Databricks IDE extension features are only available when the target deployment mode is development:

• Use the attached development cluster for bundle jobs
• Sync workspace folder files
• Select an interactive development cluster

Configure the Databricks profile for the project

When you create a Databricks project or convert a project to be a Databricks project, you configure a profile that includes authentication settings used to connect to Databricks. If you want to change the authentication profile used, click the gear icon associated with AuthType in the Configuration view.

For more information Databricks IDE extension authentication, see Set up authorization for the Databricks IDE extension.

Select compute for running code and jobs

Using the Databricks IDE extension, you can select serverless, select an existing Azure Databricks cluster, or create a new Azure Databricks cluster, for running your code and jobs. Once you have connected to compute, a cluster's ID, Databricks Runtime version, creator, state, and access mode are displayed. You can also start and stop a cluster, and navigate directly to the cluster's page details.

Tip

If you don't want to wait for the jobs cluster to start up, check Override Jobs cluster in bundle just below the cluster selection to use the selected cluster for running bundle jobs in development mode. This is not available if you are using serverless compute.

Use serverless

Serverless compute is managed by Azure Databricks. When you run workloads on serverless compute, Azure Databricks automatically allocates and manages the necessary compute resources.

1. In the Configuration view, click Select compute or the gear (Configure compute) icon.

   

2. In the Command Palette, select Serverless.

   

Use an existing cluster

If you have an existing Azure Databricks cluster that you want to use:

1. In the Configuration view, click Select compute or the gear (Configure compute) icon.

2. In the Command Palette, select the cluster that you want to use.

Create a new cluster

If you do not have an existing Azure Databricks cluster, or you want to create a new one:

1. In the Configuration view, click Select compute or the gear (Configure compute) icon.

2. In the Command Palette, click Create New Cluster.

3. When prompted to open the external website (your Azure Databricks workspace), click Open.

4. If prompted, sign in to your Azure Databricks workspace.

5. Follow the instructions to create a cluster.

   Note

   Databricks recommends that you create a Personal Compute cluster. This enables you to start running workloads immediately, minimizing compute management overhead.

6. After the cluster is created and is running, go back to Visual Studio Code.

7. In the Configuration view, next to the cluster, click the gear (Configure compute) icon.

   

   In the Command Palette, click the cluster that you want to use.

Sync your workspace folder with Databricks

You can sync the remote Databricks workspace folder associated with your Databricks project by clicking the sync icon (Start synchronization) associated with Remote Folder in the Configuration view of the Databricks extension.

Note

The Databricks IDE extension works only with workspace directories that it creates. You cannot use an existing workspace directory in your project unless it was created by the extension.

To navigate to the workspace view in Databricks, click the external link icon (Open link externally) icon associated with Remote Folder.

The extension determines the Azure Databricks workspace folder to use based on the file_path setting in the workspace mapping of the project's associated bundle configuration. See workspace.

Note

The Databricks IDE extension only performs one-way, automatic synchronization of file changes from your local Visual Studio Code project to the related workspace folder in your remote Azure Databricks workspace. The files in this remote workspace directory are intended to be transient. Do not initiate changes to these files from within your remote workspace, as these changes will not be synchronized back to your local project.

For usage details on the workspace directory sync feature for earlier versions of the Databricks IDE extension, see Select a workspace directory for the Databricks IDE extension.

Start an SSH tunnel

Important

This feature is in Beta.

You can start an SSH tunnel from the Databricks IDE extension to open a remote development session on Azure Databricks compute. This allows you to edit files and run code directly against remote compute from within Visual Studio Code.

1. To start an SSH tunnel, click the Start SSH Tunnel button in the SSH tunnel view in the extension.

2. Next, select classic or serverless compute to connect to:

   • Cluster: The cluster picker shows only clusters that use dedicated (single-user) access mode. If you already have a single-user cluster selected in your local Visual Studio Code session, that cluster is pre-selected for the SSH connection.
   • Serverless: If you do not select a cluster, the tunnel connects to serverless compute by default.

When you start a tunnel, a new Visual Studio Code instance opens in the same IDE that you are currently using. The remote session opens your home folder (users/username@databricks.com) by default and automatically displays the specific file that you were actively editing. Authentication is pre-configured, so you do not need to sign in again to connect to the remote session.

Set up your Python environment and Databricks Connect

The Python Environment section of the Configuration view enables easy setup of your Python virtual development environment and installation of Databricks Connect for running and debugging code and notebook cells. Python virtual environments make sure that your project is using compatible versions of Python and Python packages (in this case, the Databricks Connect package).

To configure the Python virtual environment for your project, in the Configuration view of the extension:

1. Click the red Activate Virtual Environment item under Python Environment.
2. In the Command Palette, select Venv or Conda.
3. Select the dependencies you want to install, if any.

To change environments, click the gear icon (Change virtual environment) associated with Active Environment.

For information about installing Databricks Connect, which enables running and debugging code and notebooks within Visual Studio Code, see Debug code using Databricks Connect for the Databricks IDE extension.