Skip to main content
Public Preview Manage the full lifecycle of Wherobots workspaces without leaving your editor. The Spatial AI Coding Assistant lets you create, start, stop, and dismiss workspaces from the sidebar, and automatically connects your local Jupyter notebooks to remote Wherobots compute with kernel selection.

​
What is a workspace?

In the Spatial AI Coding Assistant, a workspace provisions and manages a runtime in Wherobots Cloud. A runtime is a dedicated, serverless computing cluster. When you create a workspace, you are starting a runtime in the cloud that your local Jupyter notebooks can connect to and execute against.
Think of it this way:
  • A Runtime is the remote compute cluster running in Wherobots Cloud.
  • A Workspace is the editor session that wraps that runtime, giving you a local interface to manage its lifecycle and connect notebooks to it.
Creating a workspace provisions a runtime. Stopping a workspace tears down, or destroys, that runtime and halts further Spatial Unit consumption for that runtime. Most importantly, the notebook files you create in your editor stay local on your machine — the workspace simply provides the bridge to remote compute.
A workspace in the Spatial AI Coding Assistant is the local equivalent of a notebook instance in the Wherobots Cloud web console. Both provision and connect to the same underlying runtimes.
Organization Availability: Available to all Wherobots Organization Editions.

Installation & Setup

Install the Spatial AI Coding Assistant for VS Code, Cursor, Windsurf, Kiro, and other editors. Set up your API key.

AI-Assisted Notebook Development

Create Jupyter notebooks with AI assistance and connect those notebooks to remote Wherobots compute.

Submit Job Runs

Submit Python scripts as Wherobots job runs directly from your editor, with configurable runtime, region, timeout, and arguments.

​
Start a workspace

Starting a workspace provisions a runtime in Wherobots Cloud and makes its Jupyter kernels available to your local notebooks. You choose the runtime size, region, and idle timeout during creation — these determine the compute resources allocated and where they run.
1

Open the Command Palette

To open the Command Palette, click the search bar at the top of your editor and type >.Command Palette
You can also press ⌘ + Shift + P on Mac, Ctrl + Shift + P on Windows/Linux.
2

Enter the command

Type Wherobots: Create Workspace.
3

Run the command

Click Enter to run the command to create a new workspace.
4

Choose a runtime

  1. Choose the runtime configuration for your workspace, including Region, Runtime Size, and Idle Timeout.
  2. Click Create & Start.
Create Workspace runtime configuration
5

Wait for provisioning

Wait for the workspace to provision. The extension displays status updates as the workspace starts.
Once the workspace is running, the runtime is active in Wherobots Cloud and the extension automatically detects its available kernels. You can then create and connect local Jupyter notebooks to the remote runtime for execution.
Workspace startup time varies by runtime size. Larger runtimes may take several minutes to provision.

​
Pin the Extension for easy access

If the Spatial AI Coding Assistant extension is pinned to your sidebar, you can also create a new workspace by clicking the Create Workspace button.
Create Workspace button in the Wherobots sidebar
To pin the Wherobots extension to the sidebar:
1

Pin the extension

Right-click the Activity Bar (the vertical icon bar on the left side of your editor).A menu will appear showing all available extensions.
2

Select the extension

To pin the extension, find Wherobots in the list and click on it to ensure it’s checked.
Wherobots extension checked in the Activity Bar
Now you’ll have easy access to the extension in your editor’s sidebar.

​
Destroy a workspace

Stopping a workspace tears down the runtime it provisioned in Wherobots Cloud, releasing all compute resources and halting billing. Your local notebook files are unaffected — only the remote runtime is destroyed. You can start a new workspace at any time to provision a fresh runtime.
As a general rule, always make sure to save any local notebook files before ceasing working.
To stop running workspaces:
1

Open the Command Palette

To open the Command Palette, click the search bar at the top of your editor and type >.Command Palette
You can also press ⌘ + Shift + P on Mac, Ctrl + Shift + P on Windows/Linux.
2

Open the Workspaces view

Run Wherobots: Focus on Workspaces View.
3

Find the workspace

Find the workspace you want to destroy.
4

Click the stop button

To stop the workspace, do the following:
  1. Hover over the running workspace you started and a square icon () will appear.
  2. Click the square icon to stop the workspace.
    Stop (square) button icon
5

Confirm stopping the workspace

Confirm that you want to stop the workspace. Stopping a workspace tears down the remote runtime and halts compute billing. You can start a new workspace later to provision a new runtime.
As a general rule, always make sure to save any local notebook files before ceasing working.

​
Other actions

The following actions are available for each workspace in the sidebar by right clicking on a running notebook workspace:
ActionDescription
Connect KernelConnect the selected running workspace to a notebook kernel
Stop WorkspaceStop a running or starting workspace
Open in BrowserOpen the workspace in the Wherobots Cloud Console
Open Spark UIOpen the Spark UI for a running workspace (when available)
Copy Jupyter URLCopy the workspace Jupyter endpoint
DismissHide a failed workspace entry from the sidebar
You can also run Wherobots: Refresh Workspaces from the Command Palette to refresh the workspace list. Stopping a workspace tears down its runtime in Wherobots Cloud, releasing compute resources and halting billing. Your local notebook files are preserved. You can start a new workspace later to provision a fresh runtime.
As a general rule, always make sure to save any local notebook files before ceasing working.
You can also manage workspaces from the Wherobots Cloud Console.

​
Automatic MCP server configuration

The Wherobots MCP Server requires a Professional or Enterprise Organization Edition. Community Edition Organizations do not have access to the Wherobots MCP server.
The Spatial AI Coding Assistant automatically configures the Wherobots MCP Server in your editor, providing one-click setup for AI-assisted development. Once configured, you can:
  • Explore spatial data catalogs through your editor’s AI assistant
  • Generate, validate, and execute Spatial SQL queries using natural language
  • Discover dataset schemas without writing code
No manual MCP server configuration is required when using the extension. For advanced MCP server options, see Configure Wherobots MCP Server.

​
Cost considerations

Consider the following cost implications when using the Spatial AI Coding Assistant:
  • Workspaces and kernels: Each running workspace has a runtime provisioned in Wherobots Cloud that consumes Spatial Units (SUs) for as long as it is active. Stop workspaces when not in use to halt billing. Community Edition is limited to the Tiny and Micro runtimes.
  • Job runs (Professional and Enterprise only): Billed based on runtime size and duration. See Runtimes for pricing details.
  • MCP server queries (Professional and Enterprise only): SQL Sessions started by MCP queries run on a Tiny runtime and terminate after 5 minutes of inactivity. See MCP Usage Considerations.
  • Copilot usage: AI interactions count toward your GitHub Copilot usage limits. See Copilot Requests in the GitHub Documentation.

​
Troubleshooting

If you encounter issues with the Spatial AI Coding Assistant, try the following troubleshooting steps:
  • Verify the extension was installed successfully by checking the Extensions view for Spatial AI Coding Assistant by Wherobots
  • Reload your editor (Command Palette > Reload Window)
  • Reinstall the extension from the VS Code Marketplace or the Open VSX Registry (for Cursor, Windsurf, Kiro, Positron, Antigravity, VS Codium, and other editors)
  • Confirm the workspace status is running
  • Try Wherobots: Refresh Workspaces from the Command Palette and reconnect from the kernel picker
  • Ensure Jupyter integration is enabled: wherobotsjobsubmit.jupyterIntegrationEnabled
  • Review the Output panel for connection error details
  • Verify that you have a Professional or Enterprise Wherobots Organization.
    • Community Edition Organizations do not have access to job submission features.
  • Ensure the Spatial AI Coding Assistant is installed and configured with your API key.
  • Check that the script path is valid and the file is accessible.
  • Review the Output panel for error details.
  • Confirm your Organization has sufficient Spatial Unit allocation.
  • Ensure Agent mode is selected in the chat panel
  • Verify the MCP server is running (Command Palette > MCP: List Servers)
  • Confirm wherobotsjobsubmit.mcpServerEnabled is true in your editor’s Settings
  • Check that your API key is valid and re-check optional MCP runtime/region/timeout settings
  • Start a new chat session if tools are not being discovered