Christian Tzolov leads the Spring AI and MCP Java SDK projects as an R&D Software Engineer on Broadcom's Spring team. He bridges the gap between enterprise systems and AI, giving Java developers practical tools to build intelligent applications.
The secret to great AI? It's all about context—the art of context window curation. In this hands-on session, we'll build a complete AI-powered flight booking assistant for an existing/legacy booking service, combining Spring AI with a Vaadin UI.
We'll start with the basics: writing effective prompts and system instructions. Then we'll add advanced features—conversation memory, tool calling to tap into booking services, and RAG to pull answers from your documents. We'll explore how Model Context Protocol provides a unified way to interact with external resources, and dive into Spring AI's observability support for metrics, logging, and distributed tracing.
What we'll build:
- ChatClient foundation—Spring AI basics and prompt engineering
- Conversation memory—real, multi-turn conversations
- Tool calling—connecting to getBookingDetails, changeBooking, cancelBooking
- RAG integration—answering policy questions from your documents
- MCP—sharing tools and resources across platforms
- Observability—metrics, logging, and tracing
- Best practices—guardrails and security
You'll walk away with practical knowledge of integrating generative AI into Spring applications—from your first prompt to advanced features. The patterns we cover can be applied immediately to your own projects.
Searching for speaker images...