AI Agents with Java

Chapter 2. RAG as Intelligent World Model

A Note for Early Release Readers

With Early Release ebooks, you get books in their earliest form—the author’s raw and unedited content as they write—so you can take advantage of these technologies long before the official release of these titles.

This will be the 8th chapter of the final book. Please note that the GitHub repo will be made active later on.

If you’d like to be actively involved in reviewing and commenting on this draft, please reach out to the editor at [e-mail].

The previous chapter introduced vectors, embeddings, how to calculate them, and how to use them to implement semantic search. Moreover, we introduced you to different vector stores, their relationship with embeddings, and the difference between them.

In this chapter, we’ll introduce you to the Retrieval-Augmented Generation (RAG) technique. This technique lets you add context to an LLM to produce better answers to user questions by providing accurate information on a topic alongside the original user question. The core principles of RAG are the embeddings and the vector stores, so everything learned in the previous chapter is essential in this one.

We’ll cover different aspects of RAG, from simple use cases to more advanced RAG architectures. Some of the topics are:

• What RAG is and Why you’ll use it.

• Walking through a simple RAG implementation.

• Advanced Ingestion process (data privacy, security, chunking, document loader).

• Advanced Retrieval ...