How do I generate an Agent Auto Complete Evaluation?

You can generate an evaluation for a single interaction by using the API or creating an ad-hoc evaluation in the UI.

Using the API

Use the following endpoint: POST /api/v2/quality/conversations/{conversationId}/evaluations

You must provide the following information:

• conversationId
• evaluationFormId
• evaluationFormType = Automated
• agentId (optional)

Example JSON request body:

Using the UI (ad-hoc evaluation)

1. Create a new ad-hoc evaluation.
2. From the evaluation form dropdown, select an Agent Auto-Complete evaluation form.
3. Select Auto Submit and choose Virtual Supervisor as the evaluator type.
4. Click Create.

To automatically generate evaluations at scale, you must configure an Agent Scoring Rule using the AI Scoring Rules Management API.

Step 1: Create an Agent Scoring Rule

The scoring rule defines:

• Which automated evaluation form to use

• What percentage of interactions should be evaluated

• How evaluations are generated

Use the following API endpoint:

Example request:

Field descriptions

• programId – The ID of your existing Speech & Text Analytics (STA) program.
• evaluationFormContextId – The context ID of the automated evaluation form.
• samplingPercentage – The percentage of interactions that should automatically generate evaluations.
• enabled – Must be set to true to activate the rule.
• published – Must be set to true for the rule to take effect.
• submissionType – Must be set to Automated to trigger auto-complete evaluations.

Once the rule is configured and enabled, the system automatically generates evaluations for interactions that meet the defined criteria.