Epic Goal

• Develop an AI model which has the Openshift Pipelines, Tekton, and Pipelines as Code documentation in its context
• Connect customer requests to the model so that it categorizes the request as either a bug, feature request, or user error
• Additionally configure the model to respond with a brief suggestion
• Add the bot's response as a comment on the support request thread

Acceptance Criteria (Mandatory)

• Any time a customer request is made, an AI model receives the request and attempts to give a sensible response to the customer.

Dependencies (internal and external)

1. Hosting the bot will require infrastructure. Other teams have solved this using Tangerine
2. Connectivity with the bot will also require some service to turn a Slack webhook into an API request. Similar to Jira, this could be done using the Konflux support service

Open questions::

• Feedback loops: how to measure the quality of the responses

Done Checklist

• Acceptance criteria are met
• Non-functional properties of the Feature have been validated (such as performance, resource, UX, security or privacy aspects)
• User Journey automation is delivered
• Support and SRE teams are provided with enough skills to support the feature in production environment