LLMs have been trained on decades of freely available Java specifications, including JSRs, JEPs, MicroProfile and Jakarta EE. They know the patterns. They know the standards. Now let's put them to work. In this live coding session, we'll use LLM agents to build production-ready Java applications quickly. No slides, no theory - just real code, real prompts, and real results. We'll start with a typical enterprise requirement and demonstrate how to guide LLMs to generate clean, maintainable Java code following BCE/ECB architecture patterns that actually work in production. You will learn how to effectively access the LLM's in-depth knowledge of Java specifications, how to continuously improve the generated code with each iteration, and how to maintain a high velocity without creating a mess. Expect live coding, real-world scenarios from actual projects, and honest discussion about where LLMs excel and where human expertise remains crucial. Bring your questions - we'll solve them with code.

Jfokus

A conference talk transcript by Adam Bien exploring how Java's unique characteristics make it exceptionally well-suited for LLM-assisted development. Key arguments: Java's open specifications (JPA, JAX-RS, JDBC), API/SPI separation, and stable standards mean LLMs were trained on consistent, publicly available documentation, resulting in higher code quality (78% vs Python's 63%) and fewer hallucinations. The Boundary-Control-Entity (BCE) pattern provides semantic scaffolding that LLMs understand natively, enabling very short prompts to generate well-structured code. Live demos use Claude Code and Kiro to generate Nordic gods and dwarves business components in a Quarkus MicroProfile project. The talk also covers using CLAUDE.md/skills.md files to constrain LLM behavior, grounding LLMs to standard APIs only, and using multiple agents simultaneously for code generation, review, and migration tasks.

High Speed Java Development with LLMs and Agents Fast, Clean, Maintainable, by Adam Bien