Codepocalypse Now: LangChain4j vs. Spring AI
summary
Live coding builds an agentic Java app that reads calendar and attendees, generates tailored excuses, verifies them with a judge loop, and only sends emails after human confirmation; it uses LangChain4j and JetBrains Cook 1.0 plus a fake calendar, contacts and Gmail tool, plus persistent chat memory (SQLite) for six prior conversations. A lightweight classifier routes intent, a powerful model drafts text, a verifier loops until success, and a Tessel skill raises baseline correctness from ~25% by about fourfold.
AI-written summary · about
chapters (12)
- 0:00Introduction and setup
- 3:17Why agents over chatbots
- 6:50Introducing Kook framework
- 10:23Building a simple chatbot
- 13:15Adding tools to the chatbot
- 15:22Memory in agent frameworks
- 22:33Agentic workflows explained
- 28:10Demo of agentic workflow
- 35:35Visualizing agent flows
- 41:38Observability and tracing
- 44:05CooG framework and skills
- 47:18Workshop announcement
related talks
46mCodepocalypse Now Lang Chain4j vs Spring Al
02h 49mJava meets AI: Build LLM-Powered Apps with LangChain4j
51mAI-Powered Development: Hands-On Techniques for Immediate Impact
53mJava meets AI: Build LLM-Powered Apps with LangChain4j
51mJava Meets AI: Empowering Developers to Build Intelligent Apps
30mChoose Your Fighter Spring AI vs LangChain4j
01h 01mCreate #AI Infused #Java Apps with #LangChain4j
47mComparing Agentic AI Frameworks for Java
