Create Intelligent MCP based applications with Quarkus
summary
Quarkus ships a full Java MCP implementation that exposes annotated Java methods as MCP getCurrentTime tool endpoints and runs as local stdio, websockets or streamable HTTP (SSE) servers; the demo wires Quarkus as an MCP client to a Python MCP weather server and the dev UI itself becomes an MCP server — the live demo calls work end to end. Claims: supports extensions, GraalVM native builds, latest spec tasks for async tools, and a third‑party benchmark benchmarked against 15 implementations shows Quarkus near Rust and faster than Spring.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:18From prompts to intelligent agents
- 3:03Function calling interactions
- 5:27Custom integrations and MCP
- 6:30What is MCP and protocol basics
- 9:03MCP client and server capabilities
- 11:54MCP extensions and evolution
- 14:22Quarkus MCP project setup
- 17:12Quarkus with OpenAI config
- 18:37Building MCP server
- 24:20MCP server demo
- 25:48Quarkus as MCP client
- 29:19MCP servers with LLMs
- 30:23Deployment and native benefits
- 31:04Dev UI as MCP server
- 33:30Quarkus MCP details
- 34:35Performance benchmarks
- 35:37Q&A
related talks
42mMCP: Bridging LLMs with Your Applications and Data - Java/Quarkus Version
42mKeynote: Subatomic Java Microservices with Quarkus
36mCrafting Intelligent Cloud Native applications with Quarkus
51mWhy you're missing out if you're not making Quarkus extensions
49mQuarkus, the Java Stack back from the Future
48mSupersonic, Subatomic Java with Quarkus
01h 00mQuarkus from inside
46mQuarkus why, how and what