AI-Powered Speaker and Session Recommendation Engine - n8n Workflow

• Event organizers and conference planners who need rapid session curation.


• Users building conversational interfaces (like voice agents) that require instant, data-driven suggestions.


• Data scientists and automation specialists seeking advanced LLM integration n8n templates for data aggregation and analysis.


• Businesses looking to centralize speaker data using a Google Sheets n8n node.

Curation for large events is time-consuming and often based on subjective criteria. This automation solves the problem by creating an objective, AI-driven recommendation engine. By connecting event requirements (via the webhook n8n trigger), historical speaker data, and audience insights (from Google Sheets) to a sophisticated AI agent (Anthropic Claude), the system delivers optimal matches based on expertise, diversity, and audience interest. This specific n8n workflow is designed for high performance, returning a formatted response instantly to the caller while logging the results for historical tracking. Leveraging the flexibility of the n8n node system, this solution ensures scalability and accuracy in event planning.

The process begins when an external application (like a voice agent) hits the configured Webhook n8n trigger with the event details (goals, topics, audience size).

1. Request Parsing: The Parse Voice Request Code n8n node standardizes and validates the incoming JSON payload.


2. Parallel Data Fetching: Simultaneously, the workflow retrieves comprehensive speaker profiles and audience interest data from two distinct Google Sheets tabs.


3. Data Aggregation: The Aggregate All Data Code n8n node combines the request input, speaker data, and audience data into one unified JSON object, ready for AI consumption.


4. AI Analysis: This aggregated data is passed to the AI Agent n8n node, which uses the Anthropic Chat Model (Claude) to analyze the fit, generate match scores, and propose a final lineup, structured as JSON.


5. Output Preparation: The Format for Voice Response Code n8n node processes the AI's output, extracts the recommendations, structures them, and generates a plain text summary suitable for a voice agent.


6. Response & Logging: The structured response is immediately sent back to the originating system via the Send Voice Agent Response n8n node. Concurrently, the successful recommendations are logged into a third Google Sheet via the Save to Google Sheets n8n node.


7. Error Handling: The Check for Errors If n8n node monitors the AI output for parsing issues or zero recommendations, diverting any failures to an error formatting node and returning a 500 status error message to the caller.

To deploy this comprehensive n8n workflow, follow these steps:

1. Import the n8n Templates: Copy the provided JSON and import it into your n8n instance via the 'Workflows' menu -> 'New' -> 'Import from JSON'.


2. Setup Webhook: Activate the workflow. The Webhook Trigger n8n node will provide a unique testing and production URL. This URL is the entry point for your external systems.


3. Google Sheets Credentials: Configure the Google API Service Account credentials. Replace the placeholder YOURSPREADSHEETID in the Fetch Speakers Data, Fetch Audience Data, and Save to Google Sheets n8n node configurations with the actual ID of your event data spreadsheet.

1. Anthropic Credentials: Ensure your Anthropic Chat Model n8n node is configured with valid API credentials. The workflow uses Claude 4 Sonnet for advanced reasoning.


2. Customization: Review the Parse Voice Request and Format for Voice Response Code n8n nodes to adjust default input values or output text/logic if your integrated system requires different JSON formats.

Webhook Trigger (n8n trigger): The starting point, configured to listen for POST requests at the /speaker-recommendations path, enabling integration with external voice agents or scheduling platforms.
Parse Voice Request (Code n8n node): Extracts essential event parameters from the incoming payload (e.g., eventType, preferredTopics) and sets crucial metadata like requestId for tracking.
Google Sheets (Fetch Speakers Data/Fetch Audience Data): Reads detailed, multi-column data (speaker expertise, ratings, audience preferences) from specific sheet IDs and GIDs, providing the raw information required by the AI engine.
Aggregate All Data (Code n8n node): Merges the disparate inputs into a single, cohesive document, simplifying the prompt generation for the subsequent AI step in this n8n workflow.
AI Agent / Anthropic Chat Model (n8n node): The core intelligence. This Langchain integration n8n node executes the complex decision-making process, analyzing the combined data to select and score the best speaker recommendations.
Format for Voice Response (Code n8n node): A critical step that takes the AI's complex JSON output, verifies it, converts it into a structured format, and generates a user-friendly voiceResponse summary string.
Send Voice Agent Response (Respond to Webhook n8n node): Final action for successful execution, returning the optimized recommendation and summary back to the caller (HTTP 200).
Save to Google Sheets (n8n node): Logs the final recommendation JSON to a tracking sheet for auditing and long-term analytics.