Harnessing the power of vector search capabilities in Couchbase with n8n’s workflow automation platform, this tutorial guides users through creating a travel agent workflow that recommends vacation destinations based on user queries using vector embeddings for contextually relevant results. Vector search, which focuses on semantic similarity rather than exact matches, is enabled through the Couchbase Search Vector node in n8n, which allows operations like retrieval, updating, and insertion in a vector database. The process involves setting up a Couchbase Capella cluster, configuring a database with specific structures, creating search indices, and building an n8n workflow that incorporates OpenAI and Gemini for embeddings and LLM capabilities. The workflow is designed to ingest sample data and respond to user queries by converting them to vector embeddings, searching for semantically similar destinations in Couchbase, and generating responses with the help of LLM. Although the travel agent application is a demonstration, it showcases the potential of integrating Couchbase and vector search with n8n’s automation tools for diverse applications.