In this podcast episode, we explore how to create a multi-AI agent application using LangGraph and AstraDB. The app intelligently directs user queries to either a vector database (AstraDB) or a Wikipedia search, depending on the topic. We walk through the steps of setting up a vector database, embedding text data from various websites, and designing a LangGraph workflow that includes nodes for routing, vector searches, and Wikipedia lookups. Additionally, we utilize an LLM (Llama 3.1 via Groq) to make routing decisions. The end result is a versatile system that effectively manages a range of user queries by selecting the most suitable data source.
