❓Questions and Answers (Q&A)
QA Workflow
Purpose
The QA workflow allows users to connect the AI agent to their Discord or Slack channels. Users upload markdown files containing relevant information, which the agent uses to answer questions within these platforms. If the agent's confidence threshold is not met, the response is routed to a devrel (developer relations) channel for review and feedback, enabling continuous improvement of the agent's performance.
Prerequisites
A Discord or Slack account with access to the necessary channels.
Markdown files that contain data the AI agent will reference to provide responses.
Admin access to set up the devrel channel for manual review and feedback.
How it Works
This workflow consists of the following steps:
Inputs
Markdown Files: Users upload markdown files that contain FAQs, documentation, or other relevant information.
Platform Integration: The AI agent must be connected to either a Discord or Slack channel where questions will be asked.
Processing
The agent receives user questions from Discord/Slack.
It processes these queries based on the markdown files provided.
A confidence score is generated for each response. If the score meets or exceeds the set threshold, the response is posted directly in the same channel.
If the confidence score is below the threshold, the response is sent to a devrel channel for review.
Outputs
Direct Response: If the confidence threshold is met, the AI posts its answer in the Discord/Slack channel.
Manual Review: If not, the response goes to the devrel channel for approval, denial, or modification. The manual response can replace the AI-generated one, further training the agent.
Usage
Set Up the Workflow:
Connect your Discord or Slack account.
Create or identify markdown files that contain the data the agent will use.
Configure the devrel channel for manual reviews.
Upload Markdown Files:
Ensure the markdown files are structured clearly and contain accurate information relevant to the expected queries.
Execute the Workflow:
Ask a question in the configured Discord/Slack channel.
The AI agent will process the question and provide a response based on the markdown files.
Manual Review:
If the confidence threshold isn't met, the devrel team will receive a notification to approve or modify the response.
Example
Input: A user asks, "How do I reset my password?" in Discord.
Processing: The agent reviews the markdown documentation and generates a response.
Output: The response is either posted directly in the channel or sent to the devrel channel if confidence is low.
Advanced Features
Adjusting Confidence Threshold: Users can fine-tune the confidence level for when a response should be manually reviewed versus automatically posted.
Custom Workflow Combinations: The QA workflow can be combined with other workflows like document parsing or task automation to streamline operations.
Troubleshooting
Agent Not Responding: Ensure the markdown files are properly uploaded and the correct channels are configured.
Low Confidence Threshold: If many responses are routed to the devrel channel, consider adjusting the agent's training data or lowering the confidence threshold.
Log Issues: Logs for workflow activities can be accessed from the admin dashboard to track issues or performance problems.
Related Workflows
Feedback Integration Workflow: To automate the learning process based on feedback provided by devrel teams.
Data Import Workflow: For bulk uploading markdown files or integrating with external documentation tools.
Diataxis framework: [Link to the framework's documentation or reference material].
Last updated
