Evaluate Interview and Update Scores with Azure GPT-4o Mini and Google Sheets

This n8n workflow automates the process of evaluating interview responses using Azure GPT-4o Mini and updating candidate scores in a Google Sheet. It's designed to streamline the post-interview assessment process, providing an efficient way to get AI-powered feedback and integrate it directly into your candidate tracking system.

What it does

This workflow performs the following steps:

1. Triggers on Google Sheet updates: It listens for changes or new rows in a specified Google Sheet, acting as the starting point for the evaluation process.
2. Prepares data for AI evaluation: It uses an "Edit Fields (Set)" node to structure the incoming data from Google Sheets, ensuring it's in a format suitable for the AI agent.
3. Executes AI Agent for evaluation: An "AI Agent" node, configured with a specific prompt, processes the interview data.
4. Leverages Azure OpenAI Chat Model: The AI Agent utilizes an "Azure OpenAI Chat Model" (specifically GPT-4o Mini, as hinted by the directory name) to generate an evaluation and score based on the provided interview details.
5. Processes AI response: A "Code" node extracts the relevant evaluation and score from the AI agent's output.
6. Updates Google Sheet: Finally, it updates the original Google Sheet with the AI-generated evaluation and score for the corresponding candidate.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n instance: A running n8n instance.
• Google Sheets Account: With a spreadsheet containing candidate interview data (e.g., candidate name, interview responses, columns for AI evaluation and score).
• Azure OpenAI Service: Access to Azure OpenAI, specifically configured for the GPT-4o Mini model, with an API key and endpoint.
• n8n Credentials: Configured Google Sheets and Azure OpenAI credentials within your n8n instance.

Setup/Usage

1. Import the workflow: Download the provided JSON and import it into your n8n instance.
2. Configure Google Sheets Trigger:
   • Select your Google Sheets credential.
   • Specify the Spreadsheet ID and Sheet Name that contains your interview data.
   • Configure the trigger to listen for "New Row" or "Updated Row" events, depending on your use case.
3. Configure Edit Fields (Set) Node:
   • Ensure the fields being set here correctly map to the data coming from your Google Sheet and the input expected by the AI Agent. This might include fields like candidateName, interviewResponse, jobDescription, etc.
4. Configure AI Agent Node:
   • Define the System Message or Prompt that guides the AI on how to evaluate the interview. For example: "You are an expert interviewer. Evaluate the candidate's response based on the job description and provide a score from 1-10 and a brief evaluation."
   • Ensure the Input for the agent correctly references the data prepared in the "Edit Fields" node.
5. Configure Azure OpenAI Chat Model Node:
   • Select your Azure OpenAI credential.
   • Specify the Model Name (e.g., gpt-4o-mini).
   • Set any other parameters like Temperature as needed.
6. Configure Code Node:
   • Review and adjust the JavaScript code to correctly parse the output from the AI Agent node and extract the desired evaluation text and score.
7. Configure Google Sheets Node (Update):
   • Select your Google Sheets credential.
   • Specify the Spreadsheet ID and Sheet Name.
   • Set the Operation to "Update Row".
   • Map the AI-generated evaluation and score to the correct columns in your Google Sheet. You'll likely need to use a unique identifier (like a row index or candidate ID) to match the row.
8. Activate the workflow: Once configured, activate the workflow to start monitoring your Google Sheet for changes.