This n8n workflow creates an intelligent AI assistant for WhatsApp, capable of engaging natural conversations, retrieving relevant information, and maintaining persistent memory for each user. It integrates WAMM.pro for WhatsApp messaging, Google Gemini for conversational AI, Pinecone for storing conversation history and knowledge, and OpenAI for generating embeddings. The workflow begins with a webhook that receives incoming messages from WhatsApp via WAMM. It then processes these messages through a conversational agent powered by Google Gemini, which responds in a friendly, natural tone without mentioning searching activities.

The assistant uses Pinecone to retrieve previous conversation history and relevant knowledge base entries to provide contextually accurate responses. Conversations are stored back into Pinecone, creating a long-term memory for each user. The process includes data transformation and splitting for optimal storage and retrieval. This setup is suitable for automating customer support, personal assistance on WhatsApp, or managing chatbots with persistent context and a knowledge base. The workflow ensures data security by encrypting messages and organizing separate memory spaces for each user, making it scalable and extendable for various use cases.

Node Count	11 – 20 Nodes
Nodes Used	@n8n/n8n-nodes-langchain.agent, @n8n/n8n-nodes-langchain.documentDefaultDataLoader, @n8n/n8n-nodes-langchain.embeddingsOpenAi, @n8n/n8n-nodes-langchain.lmChatGoogleGemini, @n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter, @n8n/n8n-nodes-langchain.vectorStorePinecone, code, n8n-nodes-wamm.wammpro, stickyNote, webhook