Contents
JCON
▪️JCON Europe 2026: Why the Java Community Still Meets in Person
CORE JAVA
▪️Java 26 Is Boring
▪️Structured Concurrency in Java 26 (JEP 525 Deep Dive)
▪️Are the Classic Design Patterns Still Valid with Modern Java?
API & FRAMEWORKS
▪️The Gen AI Iceberg – Java Tooling Edition
▪️Building AI-Driven Java Systems with Spring AI and Java 26
▪️LangChain4j Agentic Workflows: From AI Calls to Multi-Agent Systems in Java
AI & ML
▪️Spec Driven Development: Why the Future of AI-Assisted Software Development Starts with the Specification
▪️Beyond UML: Clean Software Architecture in the Age of AI-Generated Code
▪️From Prototype to Production: Building Safe and Reliable AI Agents with Spring AI and MCP
▪️Enterprise Playbook for Generative AI Adoption
ARCHITECTURE & MICROSERVICES
▪️Hexagonal Architecture for Machine Learning: Ports, Adapters, and Model Versioning in Spring Boot
▪️Open-Core with Core Java and Vaadin — Part 1
SERVERLESS
▪️Serverless Functions Reimagined with Java 25
DEVOPS
▪️DevOps Patterns and Java 26 for On-Premises LLM Platforms in Safety-Critical Environments
IDE & TOOLS
▪️Reproducible Dev Environments Made Easy
PROJECT MANAGEMENT
▪️Code at Court – Java as Evidence
▪️RTFM – Usable Documentation
SECURITY
▪️The Whispering JAR: Java Security Lessons Hidden in a Fantasy Tale
Details
Page count: 152
Authors: Andreas Jürgensen, Artur Skowronski, Babneet Singh, Bhupendra Mishra, Cornelius May, Damiana Nascimento, Daniel Sogl, Eldar Sultanow, Johannes Bechberger, Lokesh Chenta, Loïc Magnette, Lutske de Leeuw, Marco Schulz, Mihaela Gheorghe-Roman, Mohammad-Ali A’râbi, Nikita Golovko, Richard Fichtner, Sascha Selzer, Stefan Wagenpfeil, Sven Reinck, Sven Ruppert
Editorial
From AI as a Feature to AI as Infrastructure
Artificial Intelligence has moved beyond the experimental phase.
The first wave of AI adoption focused on exploration: integrating language models into applications, experimenting with assistants, generating content, or automating isolated tasks. It was an exciting time driven by rapid innovation, curiosity, and the promise of unprecedented productivity gains.
But reality has changed the conversation.
Today, organizations are no longer asking whether AI can generate useful results. They are asking much harder questions:
- How do we operate AI systems reliably?
- How do we version and audit models?
- How do we secure AI-driven workflows?
- How do we govern prompts, embeddings, and data access?
- How do we integrate AI into long-lived enterprise systems?
- How do we build architectures that remain maintainable once the hype fades?
This issue focuses on exactly that transition: from AI experimentation to AI engineering.
Across the articles in this edition, one theme repeatedly emerges: AI is no longer just a feature. It has become an architectural concern.
Modern Java is uniquely positioned for this new reality. While other ecosystems often dominate rapid prototyping, Java continues to excel where systems must remain stable, observable, governable, and operational for years. Features introduced in recent Java releases – from Virtual Threads and Structured Concurrency to improvements in startup performance and vectorized computation – are not merely language enhancements. They are foundational building blocks for modern AI platforms and large-scale backend systems.
At the same time, the industry is learning that introducing AI into production systems also introduces entirely new responsibilities: governance, observability, reproducibility, auditability, security, and operational control.
This issue explores those challenges from multiple perspectives.
We examine how classic software architecture evolves in the presence of machine learning systems. We discuss hexagonal architectures for AI applications, model versioning strategies, retrieval-augmented generation pipelines, and on-premises LLM platforms for safety-critical environments. We look at DevOps patterns for AI systems, reproducible development environments, and the growing importance of Prompts-as-Code and semantic testing.
Beyond the purely technical dimension, we also address organizational and legal realities. AI-first cultures, enterprise adoption strategies, source code as legal evidence, software forensics, and the implications of AI-generated code are no longer theoretical concerns. They are rapidly becoming part of everyday software engineering practice.
At the same time, this issue revisits a fundamental question for the Java ecosystem itself: What is the role of Java in the AI era?
The answer emerging throughout these pages is both clear and encouraging. Java is not merely adapting to AI.
Java is becoming one of the most compelling platforms for building reliable AI systems. Not because it follows every trend, but because it provides what production systems ultimately require: clarity, structure, operational stability, scalability, and long-term maintainability.
The future of AI will not be defined solely by larger models or more autonomous agents. It will also be defined by the quality of the systems surrounding them.
That future requires software engineering. And software engineering remains one of Java’s greatest strengths.
Enjoy this issue.
