You’re a software engineer, and your project has to integrate AI. Hopefully, not just any AI. You probably need solutions that are private, efficient, and production-ready, not just a checkbox for the latest trend. Join me in this session, where we’ll apply The WHY Factor to cut through the hype and focus on what actually works.

In this hands-on session, we’ll explore how to build robust LLM applications using Java, open-source tools, and European machine learning models, ensuring compliance, security, and developer-friendly workflows. You’ll learn how to:

• Leverage open-source and European LLMs for inference, reducing dependency risks and keeping your data under your control.

• Design modular RAG architectures with Docling, enabling private, resource-efficient, and hallucination-free document processing.

• Integrate agentic workflows securely, connecting your enterprise applications and APIs without compromising privacy.

• Monitor, test, and refine your Generative AI applications locally, using observability and evaluation strategies for real-world reliability.

We’ll build a live demo to show how Java and Spring AI can help you integrate AI responsibly: local-first, open-source, and hype-free. By the end of this session, we’ll have a working application, and a bigger question: Does this AI solution actually address a need, or is it just another trend we’re chasing?

Thomas Vitale is a software engineer focused on building cloud native solutions and currently working at Systematic in Denmark. He’s the author of the books “Cloud Native Spring in Action” and “Developer Experience on Kubernetes” (co-authored with Mauricio Salatino). Thomas plays an active role in the cloud native ecosystem as a Java Champion and CNCF Ambassador. He’s the creator of the Arconia framework and a maintainer of the Docling Java project. A strong advocate of open-source collaboration, Thomas contributes to various projects in the Java and cloud-native space such as Spring AI, LangChain4j, and OpenRewrite.