Crafting Intelligent Cloud Native applications with Quarkus
summary
Quarkus integrates with langchain4j to treat LLMs as Java REST clients, showing a working demo that calls OpenAI and local Llama 3, compiles prompts from annotated interfaces, and exposes local methods as callable tools so the model invokes a getCurrentTime function and returns its result. The stack adds RAG with Postgres (SQL generation + vector search), Prometheus metrics, timeouts/fallbacks, and native compilation; all examples run live and the features demonstrably work.
AI-written summary · about
chapters (20)
- 0:00Quarkus and speaker intro
- 1:03Why AI matters today
- 3:34AI APIs as tools
- 4:38LangChain for Java in Quarkus
- 5:22Quarkus project with AI
- 6:31Quarkus Dev Mode setup
- 7:14Quarkus Dev UI and OpenAI
- 8:01Generating Java with AI
- 9:26REST clients with AI
- 10:49Testing AI endpoints
- 11:32Parameterizing AI prompts
- 13:17System prompts for AI
- 15:06AI tools calling Java
- 17:58How AI tools work
- 20:08LLMs locally with Ollama
- 21:12Metrics and fault tolerance
- 23:00Retrieval Augmented Generation
- 24:29Movie recommendation assistant
- 26:40Quarkus AI capabilities
- 28:29Q&A on AI and Quarkus
related talks
44mCrafting Intelligent Supersonic Subatomic applications with Quarkus
43mCrafting Intelligent Applications with Quarkus/LangChain4j
50mDeveloping AI applications in Java with Quarkus and Langchain4j
44mUnlocking the Future of Java: LLM Integration Unveiled by Quarkus & LangChain4j
47mQuarkus and AI Integrating Java and LLM to build intelligent applications
48mAgentic AI with Quarkus, LangChain4j and vLLM
01h 01mScaling AI with Java: Building Production-Ready LLM Applications
49mQuarkus, the Java Stack back from the Future