← Learning paths

7 lessons · take it step by step

Spring AI

Build Java AI applications with ChatClient, structured output, tools, conversation memory and a tested pgvector RAG laboratory.

Start here

Follow the order if you are new, or jump straight to a topic you need.

Start the first lesson →

  1. Spring AI: Getting Started

    20–30 min

    Run a pinned Spring Boot AI laboratory with separate fixture and Ollama modes.

  2. Spring AI ChatClient

    20–30 min

    Compose versioned requests, inspect response metadata and stream text through an injected ChatClient builder.

  3. Structured Output in Spring AI

    20–30 min

    Map model output to Java records, then validate business and evidence invariants.

  4. Spring AI Tool Calling

    20–30 min

    Expose narrow Java tools and enforce resource authorization inside the application.

  5. RAG with PostgreSQL & pgvector

    20–30 min

    Persist document vectors, filter by trusted tenant and return answers with verifiable source IDs.

  6. Advisors & Conversation Memory

    20–30 min

    Compose memory into a ChatClient without mixing users, conversations or retrieved knowledge.

  7. Testing & Evaluation for Spring AI

    20–30 min

    Separate unit behavior, provider contracts, database retrieval and live answer quality.