Build a RAG Chatbot for Movie Recommendations with Qdrant and OpenAI

Overview

This n8n workflow empowers you to build a sophisticated Retrieval-Augmented Generation (RAG) chatbot capable of providing personalized movie recommendations. It leverages the power of OpenAI's embedding models and language models, combined with Qdrant as a vector database for efficient similarity search. The workflow starts by fetching movie data from a CSV file hosted on GitHub. This data is then processed, embedded using OpenAI, and stored in Qdrant. When a user provides input, the chatbot generates embeddings for both positive and negative recommendation examples, queries the Qdrant vector store to find similar movies, and uses an AI Agent with the OpenAI Chat Model to generate a coherent and relevant movie recommendation response.