Bi-Weekly Product Release Notes - Aug 7, 2025
Release Date: August 7, 2025
Features
The following features have been added to the Aisera Gen AI Platform in this release.
Knowledge Base Article Generation
The following feature has been added to the Knowledge Base Article Generation feature in this release.
Multi-Resolution Extraction to Enhance Coverage
To improve the coverage and quality of resolution steps, we have introduced Multi-Resolution Extraction. This enhancement involves analyzing multiple representative tickets within a cluster to generate a richer and more comprehensive knowledge document.
Benefits of Multi-Resolution Extraction:
Enables visibility of multiple resolutions related to the same issue/topic.
Consolidates various resolution approaches into a single document, helping reviewers view different resolution steps at once.
Allows reviewers to edit, select, and retain only the most accurate steps.
Internal Data Process:
We have added a new data processing component called the Inlayer Analyzer, which:
Further groups tickets in sub-clusters to ensure tighter relevance around the same issue.
Selects multiple representative tickets from these refined groups.
Generates a master document incorporating multiple resolution steps.
Note: Achieving 100% tight clusters is challenging due to variations in ticket content.
To improve clustering:
We previously introduced Master and Sub-clustering to tighten the groupings (during the April 2025 release).
With the goal of incorporating more representative tickets, it is essential that these representatives address similar issues.
Hence, sub-clusters are further grouped to select more accurate representatives for document generation.
To use Multi-Resolution Extraction:
Navigate to: Content Generation → Knowledge Generation
Configure Pre-Generation Settings:
Navigate to Actions → Configurations in the Knowledge Generation window.
Under Pre-Generation Configuration, you’ll find two additional options:
Master KCS
Master Default
Note: The existing option to generate using a single representative is still available to provide greater flexibility, to generate regular documents choose KCS or Default.
Trigger KB Generation:
Select one of the above templates to trigger the generation.
The final output will contain resolution steps from multiple tickets (up to a maximum of 4).
Review & Validate:
On the side panel, you’ll see the validated chunks for each representative ticket.
Use these to compare comments and evaluate the extracted resolution steps.
Additionally, multiple Cause and Environment sections will also be available for review.
Enhancements
The following enhancements have been added to the Aisera Gen AI Platform in this release.
Connectors
The following features have been added to Connectors in this release.
Support for Confluence Offset-Based Connector
The external Confluence data source uses cursor-based pagination, and therefore our Aisera connector for Confluence also uses cursor-based pagination by default. But some of our customers use the previously supported offset-based pagination that has since been deprecated by Confluence.
If you’re using an older version of Confluence as your Data Source, you can now enable an offset-based pagination version of the Aisera Confluence connector.
To Enable Offset-Based Pagination:
Create a Confluence Data Source.
Choose the pencil icon to edit the Data Source.
Check the Enable offset-based pagination box in the Configuration tab of the Data Source details.
Run the Data Source.
Ask your Aisera Team to tail the conn-ds pod log and verify that the connector requests are paginated using start and limit.
Agent Assist
The following features have been added to the Agent Assist feature in this release.
Default Values Shown for NBA Generation from Messages Job
The default values for the NBA Generation from Messages System Job that is used with the Agent Assist applications/bot now displays the default values in the System Job creation fields.
Each field now contains the default values or value syntax as a background image in the field.
Field Definitions for NBA from Messages Job:
Data Source IDs: Enter one or multiple Data Source IDs (comma-separated). Leave this field blank if you want to include all data sources associated with that bot.
Channel Names: Specify one or more channel names (as they exist on your Aisera tenant (comma-separated), or leave the field empty to include all channels.
Examples: support-hr-compliance, ask_hr_person.
You can find Channel Names in the Overrides tab, after you click the pencil icon to edit the Data Source:
Closed Thread Interval: This helps decide which threads to use for NBA creation — acting as a proxy for closed tickets.
The default is 30, meaning threads last updated 30 days or before are considered. If you're working with newly ingested data (for instance, ingested today), set it to 0 or leave the field blank to include the latest threads.
Thread Limit: This controls how many messages to ingest. Leave it blank to ingest all. Use this if you want to limit the scope, for example, when testing with a few hundred or a thousand messages.
Hover Text for NBA from Messages Job
Each field also contains hover text (descriptions that appear when you move the computer mouse over a field).
Scheduling for NBA Generation from Tickets or Messages Jobs
You can now set the System Jobs schedule for NBA Generation from Tickets or NBA Generation from Messages in the Agent Assist window of the Aisera Admin UI by navigating to Agent Assist → Configuration.
Two new buttons with the titles, Start NBA Generation Job and Start NBA from Messages Job are now in the upper-right corner of the Agent Assist Configuration window, next to the Start Knowledge Predictions Accuracy Job.
When you select one of these buttons, a New NBA Generation Job or New NBA from Messages Job window appears. The default options for each of the fields appear as background text within the field.
Hover text for each field gives you a description of the field.
The schedule will run for the Bot for the set of Data Sources specified in the Data Source Ids field.
Notes:
A schedule is set for Tickets and/or Messages.
Only one schedule can be set for Tickets and/or Messages for one Bot.
To clear the schedule, stop the schedule, or update an existing schedule, set the start date and time to now and choose OK.
To add a new data source to the recurring schedule, run the NBA Generation Job manually and update the list of data sources (same if a new slack channel needs to be added).
To see the current schedule settings, click one of the Start NBA Generation buttons and the New NBA Generation Job window will show the current schedule (this is the same for messages).
Last updated