Pure Java Enterprise AI/LLM Integration (EAI 2.0)
summary
Shows concrete Java 21 patterns for enterprise LLM use: Quarkus/MicroProfile services call Claude, ChatGPT, Llama and AWS Bedrock via Java 11 HTTP client or MicroProfile REST client, or run local models with llama3-java packaged as a single executable jar (copy to S3, java -jar, or run on EC2). Demonstrates a sentiment analyzer that downloads and caches the model, answers in ~2s, and passes parameterized unit tests. Highlights cost traps (RAG calls can cost ~$7 each, Lambda waiting drove a €1,400 bill) and recommends langchain4j for RAG, caching, retries and observability.
AI-written summary · about
chapters (12)
- 0:00Introduction and background
- 2:59Java and cloud infrastructure
- 5:03Enterprise challenges with LLMs
- 13:02Core principles for enterprise AI
- 18:45API patterns and communication
- 23:28Operational challenges of AI
- 29:23Local and offline LLM execution
- 33:41Infrastructure and architecture
- 36:28Code demonstrations and Jlama
- 42:28Sentiment analysis implementation
- 48:30Design patterns and conclusions
- 56:31Session wrap-up
related talks
52mLivecoding: Creating Beautiful Java Code With LLM and Agents
01h 01mScaling AI with Java: Building Production-Ready LLM Applications
48mAgentic AI with Quarkus, LangChain4j and vLLM
01h 02mJava + LLMs (Build LLM Apps in Java with JakartaEE)
47mQuarkus and AI Integrating Java and LLM to build intelligent applications
36mCrafting Intelligent Cloud Native applications with Quarkus
32mScaling Generative AI: Building Production-Ready LLM Applications
45mComplementing LLM with agentic AI: a Quarkus story on AI