Building AI Agents with Spring & MCP
summary
Demo builds an adoption chatbot in Java 25 using Spring AI 1.1, wired to Amazon Bedrock models and a PG vector Postgres store. It implements RAG with embeddings (Bedrock/cohere model), rolling chat memory persisted to the DB, a question-answer advisor for similarity search, and agent tools for scheduling. The local tool is refactored into an MCP server and the system makes a verified remote MCP schedule call end to end, with observability via Spring Actuator and Micrometer.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 0:51Why Java for AI agents
- 3:38Choosing AI models
- 4:41Setting up the project
- 7:09Configuration basics
- 7:50Building the chatbot
- 9:13Adding memory to chat
- 12:18User-specific memory
- 13:00Giving the agent a mission
- 14:46Introducing RAG
- 16:10Vectorizing dog data
- 17:11Structured data from AI
- 18:13Building tools for adoption
- 20:39Refactoring to MCP
- 21:43Connecting MCP server
- 22:46Testing MCP integration
- 23:49Observing token usage
- 25:11Summary and next steps
related talks
49mBuilding AI Agents with Spring & MCP
40mBuilding AI Agents with Spring & MCP
54mFrom Single-Shot LLMs to Intelligent Agents: Building Scalable AI Systems with Spring AI and MCP
01h 00mSpring AI
44mThe Spring AI Ecosystem in 2026: From Foundations to Agents
52mBeyond local tools: Deep dive into MCP with Spring AI
32mFrom Monolith to AI Agent Modernizing Java Systems with MCP
41mFrom Chat to RAG to MCP: Enhancing Java Applications with AI