Skip to main content

v1.0.229

Date released: February 12, 2026

New features and enhancements​

AI agents​

SQL agent​

Previously, saving a workbook would end the active session, requiring users to manually navigate back to start a pipeline. Users can now immediately Create a new pipeline or Add to an existing pipeline directly after saving a workbook, without leaving the page.

Sharing agent thread​

Redesigned the sharing UI to mirror Google Drive’s General Access model, enabling one-click public access via an Anyone with the link drop-down.

*sharing agents publicly

Improved AI generation interface​

Relocated AI Generate function from the bottom bar into a dedicated two-tab layout (Settings and AI Generate) to reduce UI clutter and increase feature discoverability.

*ai-generate-new-tab

Onboarding​

Implemented a maximum file size limit for onboarding uploads to ensure faster processing, smoother previews, and a more responsive setup experience. Users now receive immediate, helpful feedback if a file exceeds the limit, preventing failed uploads across all supported formats (Excel, CSV, PDF).

Orchestration​

The Edit button is disabled while an orchestration is Running status. This eliminates the risk of mid-run errors or data mismatches and workflows run exactly as planned, every time.

Bug fixes​

BI​

  • Issue: Users encountered errors when attempting to edit BI queries connected to Redshift. Specifically, the connection type would default to a blank value, triggering a "Spark error" and preventing the query from running.

  • Fix: We resolved a synchronization issue to ensure that the JDBC provider type remains correctly mapped when a workbook is opened for editing.

  • Impact: This restores the ability to modify and test Redshift-based queries without manual reconfiguration or system errors, ensuring your data reporting remains agile and accurate.

Column extraction​

  • Issue: Fixed an issue where Extraction Flows would remain stuck in a Running status indefinitely if they encountered a duplicate column error.

  • Fix: The system now correctly detects duplicate column failures and updates the flow status to Failed.

  • Impact: This resolves an issue where the Run button remained disabled, preventing users from initiating subsequent extractions.

Platform​

  • Issue: When a new user was invited but had not yet completed their account setup, their name appeared as undefined undefined across various pages in the platform.

  • Fix: Updated how the system handles pending accounts. The interface will now correctly display the user’s email address or a placeholder name until they complete their profile registration.

  • Impact: This ensures that the team dashboards and user lists remain clean, professional, and easy to read, even while waiting for new members to join.

Was this helpful?