Skip to main content
Version: 4.0.1

Release notes

Version 4.0.1

About Release
  • Released: February 13, 2025
  • Compatible platform version: IA Cloud 10.2.9+

As part of the 4.0.1 Kayla release, we introduce a configuration to integrate with custom data providers, Java Native Workers to replace WebHarvest dependencies, and a schema-based approach to handling data in Business Processes. It also features improvements to RPA connectors and messaging and warnings in the user interface.

New features

  • We deprecated WebHarvest dependencies and migrated the AI Agent flows to Java Native Workers (JNW), except for some RPA connectors that are not compatible with JNWs yet. The move offers a significant enhancement in the execution performance of Java-based tasks. The execution time of a JNW task is 80-90% less than that of a WebHarvest one.

  • You can now set Kayla to source input from custom data providers by configuring those on the new Data providers step in the AI Agent's configuration wizard. The new capability allows you to integrate Kayla seamlessly with your internal systems without complicated workarounds or much engineering effort.

    To support the new capability, we introduced the Custom connector Business Process (BP). It is available as part of the default installation together with the existing File ingestion and Requirements validator BPs. Changes were also made to the flow of the Requirements validator BP: it got a new Custom connector step making synchronous calls to the Custom connector BP to ingest data from configured data provider integrations (if any).

  • All out-of-the-box BPs and any custom BPs you might build for Kayla are now using the schema-based approach to handle inputs and outputs. Now, each step in a Business Process can have a contract defining variables expected as input for this step and those produced as its output. The new approach significantly reduces the amount of data passed across steps, thus increasing the overall automation performance.

Improvements

  • The BCB modules within RPA connectors were refactored to improve the organizational structure.

  • The messaging mechanism for Kayla was improved to reduce the overall message volume and make more precise descriptions of what happened. Now, Kayla sends one warning message for missing field values per entity and one error message for missing required fields per entity.

  • The logic for statuses is simplified so that you can have all required information handy in one place. Status constants are removed and a new enum is created that contains both Kayla's Business Process status and the Manual Task label.