Gmail Customer Support Auto-Responder with Ollama LLM & Pinecone RAG

This workflow triggers on new Gmail emails every minute, classifies intent with a LangChain Text Classifier to filter support queries, and skips irrelevant ones. It then leverages an Ollama-hosted LLM AI Agent to generate warm, personalized replies signed by 'Mr. Aashit Sharma from Setidure Technologies,' enhanced by real-time FAQ retrieval from Pinecone vector database for accurate, context-aware information.

After crafting the response, it labels the email in Gmail (e.g., 'Handled' or 'Auto-Replied') for easy tracking and sends the formatted reply back to the customer. Ideal for businesses in crypto, blockchain, or any sector needing scalable support automation.

Benefits include instant responses reducing response times from hours to minutes, cost savings by minimizing human agent involvement, consistent brand-aligned tone, and improved customer satisfaction through precise, knowledge-based answers. Use cases: portfolio tracking support, crypto trading queries, general customer service in high-volume inboxes. Requires self-hosted n8n with community nodes.