Real-World AI Patterns with Spring AI and Vaadin
summary
Live demos show a Vaadin chat UI wired to Spring AI 1.00 and GPT41, with a message memory advisor that reattaches recent turns, token streaming via Reactor (add Spring WebFlux), and system prompts for tone. Guardrails run as input/output advisors (example blocks phone numbers, can run locally on Llama), RAG uses embeddings stored in a PostgreSQL vector store for semantic retrieval, and tools/agents are exposed via annotated service methods and MCP; a filesystem-driven theme generator is built and works but returns inconsistent themes.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:45Spring AI with Vaadin setup
- 3:09AI model configuration
- 5:15Basic chat interaction
- 7:23Memory for AI conversations
- 8:30Custom AI system prompts
- 9:34Streaming AI responses
- 11:20AI safety guardrails
- 18:55AI actions beyond chat
- 23:12Retrieval augmented generation
- 27:02RAG with vector stores
- 32:21Multimodal AI inputs
- 35:32AI tools and agents
- 40:54Live demo: AI-driven UI
- 43:44Summary
related talks
49mPractical AI in Java
01h 00mSpring AI
51mIntelligent Applications with Spring AI
52mIntelligent Beans with Spring AI
27mBuilding AI Agents with Spring & MCP
49mBuilding AI Agents with Spring & MCP
35mUnleashing Creativity with Spring AI
49mSpring AI: Seamlessly Integrating AI into Your Enterprise Java Applications