Enthusiast: The Open-Source Toolkit for Building RAG-Powered AI Agents for E-Commerce Workflows

Traditional e-commerce chatbots often fail due to generic responses and inability to handle complex data. Enthusiast, an open-source framework by UpsideLab, addresses these issues by using RAG and LLMs. It creates a unified, searchable knowledge platform that understands product language. The system connects to existing e-commerce systems, turning scattered data into a central interface. Teams can build AI agents for tasks such as customer support, content creation, and knowledge management. Enthusiast is self-hostable, allowing control over data and models. RAG enables contextual search, providing factual responses grounded in product data. The framework includes pre-built agents for customer support which can answer customer queries and marketing content to generate ads and newsletters. It also offers internal knowledge management and search to unify scattered information, and product recommendations utilizing open-ended queries for better product discovery. Finally, content verification helps ensure accuracy, making Enthusiast a flexible AI solution for e-commerce.