Training

5min

Training AI Agents with Enjo

The Training feature in Enjo allows AI Agent administrators to enhance the performance of their AI Agents by refining responses to past user interactions. By addressing unanswered queries or improving suboptimal responses, administrators ensure the AI remains accurate, consistent, and helpful over time.



How it works

1. Access the Training Tab:

  • Navigate to the Training tab located above the "Insights" section in your Enjo dashboard.
Document image


2. Select an AI Agent:

  • Use the dropdown menu near the page title to choose the AI Agent you wish to train.
Document image


3. Explore the Conversations Table:

  • The table displays past user interactions with key fields:
    • Date: The date of the interaction.
    • User Condensed Query: The AI Agent's interpreted version of the query based on context.
    • AI Agent Response: The response provided by the AI Agent.
    • Training Actions: Options to refine the AI's responses.

4. View Additional Context: Click the View Details button to open the details drawer, which includes:

  • Message Details: Important metadata about the message.
  • Attachments: View any attachments associated with the query.
  • Generated Query: How the AI interpreted the user's input.
  • Original Query: The user’s actual message.
  • Standalone Query: The AI Agent's context-free interpretation of the query.
  • AI Agent Response: The initial response provided by the AI Agent.
  • Training Actions: Available options for improvement.
Document image


5. Apply Training Actions: Choose one of the following actions to enhance the AI Agent’s performance:

  • Add Answer: Add the query and response as a new standalone entry in Enjo Answers.
  • Use Answer: Link the query to an existing answer by adding it as a question variation.
  • Use AI Action: Associate the query with a predefined AI Action for automated workflows.
  • Use Request Form: Map the query to an existing request form for structured responses.
  • 
Document image

  • Conversation Review: Quickly identify and address queries needing refinement by browsing past interactions.
  • Targeted Training Actions: Enhance the AI’s knowledge by adding new answers or linking queries to existing solutions.
  • Customizable Filters: Focus on specific types of interactions, such as unanswered queries or a specific date range.
  • Details Drawer: Access comprehensive conversation context for informed decision-making during training.



Updated 09 Dec 2024
Doc contributor
Did this page help you?
PREVIOUS
Help Desk Assessment
NEXT
Enjo API
Docs powered by Archbee