Skip to main content

​
Export Inflection Data to Databricks

For information on tables and schemas that inflection exports to Databricks, please read this
These steps need to be run by someone with access to
  1. Go to Connections → Data Connections, then click Add new connection. Add new connection button on the Data Connections page
  2. Choose your platform: Select Databricks as your destination platform.
  3. Select connection type: Choose Export Data to Databricks. Choosing Databricks as the platform and Export Data to Databricks as the connection type
  4. Enter your DW credentials Give a name for your data warehouse connection and fill in your credentials.
    • For Databricks: Enter client ID, client secret, host, catalog, schema and HTTP path.
Make sure the service principal used for Inflection access has at least the following minimum permissions on the target catalog for data export.Minimum permissions required on the target catalog for the Inflection service principal
  1. Click Create Connection to set up the connection. Note that syncing will NOT BEGIN and connection will be in INACTIVE state until the credentials and destination tables are validated by your Inflection team.
We provide a custom SQL script tailored to your configuration. You’ll need to run this script to create the export tables in your DW.Sample SQL. Do not use this directly. Please refer to the SQL shown on inflection UI for your connectionThe script creates tables with the necessary structure and permissions (INSERT, UPDATE, DELETE) for Inflection to write data.
Once your credentials are saved and your data team has created the required tables in your warehouse, your setup is complete.At this point:
  • Inflection’s team will validate the setup and activate the sync for you.
  • No additional steps are required from your side.
    • Optionally, you may test or delete the connection at this stage
  • You will be notified once the sync has started and data is being sent.
Activation may take up to two business days. Please ensure your data team notifies you once your tables are set up in your data warehouse.
And that’s it! Once you are set up, you can monitor sync health by following our guide here.

​
Import Databricks Data to Inflection

Inflection can read data directly from your Databricks workspace using Databricks-to-Databricks Delta Sharing. You stay fully in control of what’s shared; Inflection only receives read-only access to the specific tables you choose. This instructions below walk you through:
  1. Setting up a Delta Sharing share in your Databricks Unity Catalog.
  2. Granting Inflection access using our metastore identifier.
  3. Configuring the Databricks connection inside Inflection.
To complete this setup, you’ll need:
  • A Unity Catalog–enabled Databricks workspace.
  • Permissions to:
    • Create and manage shares and recipients in Unity Catalog.
    • Grant SELECT access on the tables/schemas you want to share.
  • The Inflection Databricks metastore identifier:

​
Step 1: Choose what to share

Decide which objects Inflection should have access to:
  • One or more tables, for example:
    • your_catalog.your_schema.product_activity
    • your_catalog.your_schema.contacts
  • Or an entire schema in a catalog (e.g. your_catalog.your_schema).
You can add or remove objects from the share later; no need to get it perfect on day one.

​
Step 2: Create a Share in Unity Catalog

You can create the share using the Catalog Explorer.
  1. In Databricks, open the Catalog panel.
  2. Click the gear icon (⚙) and choose Delta Sharing.
  3. Under Shared by me, click Create share / Share data.
  4. Enter a Share name, for example:
  5. Click Save and continue.
  6. On the next step, add the tables/schemas you want to share to Inflection, then save.
You’ll use this exact Share name again when you configure the connection in Inflection.

​
Step 3: Create a Databricks Recipient for Inflection

Next, create a recipient that represents Inflection’s Unity Catalog metastore, using the identifier we provided.
  1. In Databricks, go to Catalog → ⚙ Delta Sharing.
  2. Go to the Shared by me tab.
  3. Click New recipient.
  4. Choose a Recipient name, for example:
  5. For Recipient type, select Databricks.
  6. In the Databricks metastore identifier field, paste:
  7. Save the recipient.

​
Step 4: Grant the Share to Inflection

Finally, grant the share to the recipient you just created.
  1. In Catalog → ⚙ Delta Sharing → Shared by me, click your share (inflection_share).
  2. Open the Recipients tab.
  3. Add inflection_recipient and grant it access.

​
Step 5: Open the Databricks Connector in Inflection

  1. Log in to your Inflection workspace.
  2. Navigate to Connections → Data Connections.
  3. Click + / Add connection to open the Connection Marketplace.
  4. Under Data Warehouse, choose Databricks. Databricks tile under Data Warehouse in the Connection Marketplace

​
Step 6: Choose sync type

On the “Setting up Databricks connection” screen:
  1. Select Read Databricks Data
  2. Click Enter connection details. Read Databricks Data option on the Setting up Databricks connection screen Databricks connection details form

​
Step 7: Mapping DW to Inflection Objects

Use the document below to Map Inflection Tables to Databricks:the related Help Center article
And that’s it! Once you save the mapping and you connection is active, you can monitor the sync health. If you encounter any issues, reach out to the Inflection team.