Create #AI Infused #Java Apps with #LangChain4j
summary
Shows a working Quarkus + LangChain4j stack that lets Java apps call LLMs via @RegisterAIService annotation, keep conversation state with chat memory and memory IDs, and invoke deterministic code with function-calling tools. Demonstrates RAG using embeddings stored in PGVector vector store (ingest + retrieval), local inference options (Ollama/Podman/jLlama), retries/fallbacks and OpenTelemetry for observability, and input/output guardrails that stop prompt injection—demo chatbot obeys a 11‑day cancellation policy and blocks the injected "delete DB" command.
AI-written summary · about
chapters (14)
- 0:00Introduction and setup
- 1:56AI landscape for Java developers
- 4:07LangChain4j overview and setup
- 6:37Prompting and system messages
- 10:50Stateful conversations with memory
- 14:05Function calling and tools
- 18:54Guardrails for AI safety
- 25:31Retrieval Augmented Generation basics
- 31:41Agentic AI and MCP introduction
- 37:41Demo setup and configuration
- 41:14Live demo of AI customer support
- 55:13Fault tolerance and observability
- 58:05Model Context Protocol overview
- 1:00:12Conclusion and Q&A
related talks
43mCrafting Intelligent Applications with Quarkus/LangChain4j
44mUnlocking the Future of Java: LLM Integration Unveiled by Quarkus & LangChain4j
53mJava meets AI: Build LLM-Powered Apps with LangChain4j
50mDeveloping AI applications in Java with Quarkus and Langchain4j
47mCreating AI-infused Apps, the Easy Way 🤙
02h 49mJava meets AI: Build LLM-Powered Apps with LangChain4j
01h 01mScaling AI with Java: Building Production-Ready LLM Applications
47mJava Meets AI: Integrating LLMs into Your Applications with LangChain4j