Import BI files using Genie Code

Beta

This feature is in Beta. Workspace admins can control access to this feature from the Previews page. See Manage Databricks previews.

Add a Tableau or Power BI file to Genie Code and it builds an AI/BI dashboard that replicates your visualizations, connected to metric views that mirror the underlying business logic.

Requirements

• Partner-powered AI features enabled for both the account and workspace. See Partner-powered AI features.
• The Import from External BI to AI/BI Beta enabled in your workspace. See Manage Databricks previews.

Upload a file

You can upload a file (.twb, .twbx, .tds, .tdsx, or .pbit) directly into the Genie Code chat, or reference a file stored in a Unity Catalog volume. For help exporting files, see Tableau export documentation or Power BI report templates.

Upload directly to the chat

1. Open an AI/BI dashboard.
2. Click the Genie Code icon in the top-right corner to open Genie Code.
3. Select Agent mode.
4. Use the /importBI command to begin importing.
5. Attach your file from Tableau or Power BI to the chat.

note

Direct file uploads have a 100 MB size limit. For larger files, upload your file to a Unity Catalog volume first, then reference it inline. You can also unzip a .twbx file and upload the extracted .twb file instead.

Reference a file in a volume

If your file is already stored in Unity Catalog, use the /importBI command followed by the volume path in the Genie Code chat:

/importBI

@/Volumes/my_catalog/my_schema/my_volume/sales_workbook.twb

Promote metric views to Unity Catalog

Genie Code creates metric views for the measures and dimensions in your file. By default, these are local to the dashboard, which works well for iterating. To share or productionize them, you must promote them to Unity Catalog.

Promoted metric views are reusable across dashboards, Genie Spaces, and notebooks, and get full Unity Catalog governance: access controls, lineage, and discoverability. Local metric views can only be referenced within the dashboard where they were created. Databricks recommends promoting metric views to Unity Catalog.

For instructions, see Export to a Unity Catalog metric view.