Release Notes: 2025

Write transformation outputs to new database tables without leaving your workflow.

In addition to writing to existing tables, the Database output step can now create tables in connected external databases and automatically import them to your Catalog as new items.

When creating a new catalog item, specify:

For details, see Database Output Step.

We’re introducing fixed maintenance windows to deliver critical updates while minimizing disruption to the platform’s uptime and availability.

This allows us to perform operations that require brief downtime, such as upgrading authentication services, databases, and message brokers, implementing complex metadata model enhancements, and migrating major components.

Export data from transformation plans directly to file storage locations such as Amazon S3.

The new File export step enables you to:

Common use cases include:

When used with reference data tables, scheduled exports always use the latest published version at the time the plan runs.

For details, see File Export Step.

You can now change the data type of attributes in reference data tables and automatically convert existing data through the Data Structure tab.

The conversion preview shows original and converted values side by side, helping you identify data that cannot be converted and will be deleted.

For details, see Change data types.

Save frequently used prompts directly from AI Agent conversations to your personal library. Reuse them for faster, consistent results on repeated tasks.

Test transformation expressions and steps with sample data before applying them to your full dataset. Instantly preview how expressions or steps modify input data without running a full plan preview.

This is especially valuable for validating AI-generated configurations before deployment.

Test with confidence using:

For details, see Test Expressions.

Set up recurring execution of your data transformations using simple interval scheduling or advanced cron expressions. Run transformations automatically on your preferred schedule without manual intervention.

Configure schedules with flexibility:

For details, see Schedule Transformation Plans.

Set thresholds for DQ rules to add context to your data quality metrics. This way, you can clearly determine which results indicate acceptable data quality and which require critical attention, making your DQ results truly actionable.

Leverage thresholds to:

Thresholds can be set as static values at multiple levels:

Once configured and evaluated, you’ll find the enriched DQ status displayed across the platform:

Streamline your reference data approval process: review and approve multiple records simultaneously instead of handling them one by one.

All records pending review are now consolidated into a single subtask for easier management. Select and approve multiple records with one action, letting you focus on decision-making rather than routine tasks.

For step-by-step instructions, see Approve and publish changes.

Find specific records faster with new status-based filtering. Display only the records that need your attention without sifting through irrelevant data.

You can filter only:

For details about filtering, see Filter records.

Adjust the structure of your dataset effortlessly with the new Edit Schema step in the plan editor. Reorder columns or rename attributes in a single action, reducing the time spent on manual schema adjustments.

Previously, schema modifications required multiple steps:

With the Edit Schema step, you can now:

For details, see Edit Schema Step.

Connect your existing Identity Provider (like Entra ID or Auth0) for single sign-on (SSO) access to Ataccama ONE - Agentic Data Trust Platform. Configure SSO directly through your tenant’s Cloud Portal settings.

The integration supports both OIDC and SAML protocols and automatically imports user groups from your Identity Provider, allowing you to set up application permissions at scale rather than individual user management.