Zero to RAG in 20 minutes

​Agentic AI is moving fast, but getting a first application off the ground still trips up many teams. Which models should you use? Where should your data live? Should you build it yourself or buy a platform?

​On November 19, we will try to answer those questions and show you how to build agentic AI applications that can go from prototype to production without a rewrite. We’ll walk through how vector databases work under the hood, how a dedicated vector database stacks up against extensions like pgvector, and how to decide which fits your use case.

​You’ll also learn about the Weaviate Query Agent, which handles natural language querying over your data with no custom pipeline required. In a live demo, we will load a dataset in Weaviate and use the Query Agent to build a complete agentic AI search app in about 20 minutes.