Databricks Integration

The Databricks integration syncs candidate data between Great Question and your Databricks warehouse. Setup has two parts: connecting Great Question as an application in Databricks, then entering your credentials in Great Question.

​

Part 1: Set up Databricks

​

Option A: Partner OAuth application (recommended)

Enable Great Question as a partner OAuth application in Databricks:

1. In the Databricks account console, go to Settings.
2. On the App Connections page, click Add Connection.
3. Enter the application name as GreatQuestion.
4. Add the redirect URLs and set the SQL access scope.
5. Generate a client secret. Databricks issues a 60-minute access token and a 10080-minute (7-day) refresh token by default.
6. Save, then copy your Client ID and Client Secret.

​

Option B: Service principal user account

This option is currently unavailable. If you’d prefer this approach, contact support and we’ll let you know when it’s restored.

​

Part 2: Configuration

As an admin in Great Question, go to Settings > Integrations, find Databricks, and enter:

• Host URL (for example, https://dbc-123456.cloud.databricks.com/)
• Client ID
• Client Secret

Click Connect. After a successful connection, you should see a screen that looks like the image below: Still in Great Question, enter the following details from your Databricks workspace:

• Warehouse ID (copy it from the SQL Warehouses section in Databricks)

• Catalog (defaults to workspace if left blank)
• Schema (defaults to default if left blank)
• Table name for importing candidates, and a separate push table name if you’re exporting candidates back to Databricks (defaults to gq_candidates)

For example, if your setup looks like this in the Catalog Explorer: Your information would be: Catalog - workspace Schema - default Table - brix_candidates

​

Enable syncing

Once connected:

• Check Enable pull candidates to import candidates from Databricks on a daily schedule.
• Check Enable push candidates to export candidates back to Databricks on a daily schedule, in addition to running an export on demand from the Candidates page.
• Add a gq_import boolean column to your table and enable import filtering to control which records are pulled in.
• Choose what happens when a record disappears from your table: unsubscribe the candidate, delete them, or do nothing.

​

Custom attributes

All table columns become available once you create custom candidate attributes in Great Question that match your Databricks column names exactly. Pushed data can use different field names, configured per attribute.

​

Sync frequency

The integration uses OAuth2 authentication and the Databricks REST API with SSL encryption for secure data transfer.

---

Still need help? Contact us at support@greatquestion.co — median response time is 19 minutes during support hours.