Helidon and AI - Building Intelligent Applications
summary
Demonstrates building AI services with Helidon 4.2 and a new build‑time DI called Helidon Inject that generates reflection‑free Java code and a LangChain4J integration. Shows live code: an AI chat service, a tool function returning current time, and provider swapping (OpenAI demo token → Llama/Colama) without changing business code. Reports packaging wins: jlink image ~74% smaller (≈88 MB), native image ~88% smaller (≈41 MB) and up to 40x startup with native images and 30x startup with CRaC snapshots; OCI GenAI and embedding store integration also works.
AI-written summary · about
chapters (8)
- 0:00Introduction to Helidon
- 1:46Helidon architecture and features
- 10:41Using AI with LangChain4j
- 16:36New features in Helidon 4.2
- 22:22Helidon LangChain4j integration
- 27:36Building an AI application demo
- 39:54OCI GenAI integration
- 42:41Conclusion and Q&A
related talks
31mFly like a rocket with Helidon
29mFly like a rocket with Helidon
01h 03mBuilding AI powered Applications with Java
47mFly like a rocket with Helidon
53mUnlocking AI's Potential: Smarter Coding in Apache NetBeans IDE
45mJakarta EE Meets AI
55mAI in Java? Meet LangChain4j and Build Intelligent Services
45mAI-infused command-line applications