Generative AI and its applications are no longer just a prerogative of data scientists using Python but, with Java and Spring Boot skills, it could be approached through the new "Spring AI" framework.
We will see step-by-step how we can implement a Spring Boot application based on Retrieval Augmented Generation (RAG) using a knowledge base stored in a VectorDB and MCP Agents for structured data, leveraging public, hybrid or on-premises LLMs, showing tips&tricks in ingestion, choosing the LLMs models and prompt engineering. An assisted approach, where most of the moving parts will be tested before being implemented will be also shown, to quickly generate code reflecting the tested configurations and use it as a baseline for an AI Microservice. The transition phase to production, through a Backend platform, will not be overlooked.

Accepted @ JavaCro'25