Workshops

Almost 8 years have passed since the initial release of JUnit 5. But a lot has happened since then! In 12 additional feature releases the JUnit team has continually enhanced the expressiveness of the testing framework and the flexibility of the platform. This workshop is for everyone who is interested in a introduction to JUnit or who wants to catch up with the latest features and enhancements by doing hands-on exercises.

We will start with the basics, explain the lifecycle of tests and which possibilities exist for influencing it. Assertions, display names, tags, order and timeouts will be additional topics. We will then focus on parameterized tests, including custom argument sources. Last but not least, we are going to go into detail on JUnit Jupiter's extension model, introduce its core extensions, and write custom extensions together.

In this workshop, you will build an AI-powered chat assistant for a coffee shop using Helidon and LangChain4J. The assistant will understand natural language, provide menu recommendations, and guide users through the ordering process.

After numerous great sessions about EclipseStore, it’s finally time for code. In this workshop, you will learn setting up a new EclipseStore in-memory database instance and associated storage, storing and loading objects, using lazy-loading, searching and filtering with Java Streams API, how to handle class changes by using legacy type mapping, deploying and running EclipseStore applications in production. Additionally, you will get many useful tips such as how to design a suited object model, different configurations, designing object models, and avoiding most common pitfalls.

This workshop will show you how easy it is to create and use Vector embedding using JDBC in your Java application. Hands on lab to explore the new Oracle Database capabilities, the developers will gradually build a GenAI app by learning the basic concepts and then add features to implement a full-fledged intelligent application.

Local AI models are becoming increasingly important in modern software development, especially in areas that require offline-capable, data-protection-friendly, and powerful systems. In this workshop, we examine how local AI models with Ollama and Retrieval Augmented Generation (RAG) can be integrated into existing Java applications to carry out data-driven optimizations.

The open-source project "Expedition Planner", a Java-based web application for planning and organizing expeditions, serves as the playground. The workshop aims to develop an AI-based optimization component that analyzes and automatically improves expedition tasks and equipment lists using RAG. Technologies such as Langchain4j are used to integrate Ollama models and RAG seamlessly.

At the end of the workshop, the participants understand how to integrate and use local AI models and RAG in Java-based applications. We develop our AI expansion for the expedition Planner, offering data-based optimization options. In addition, we learn how LLMS can be used productively within a closed infrastructure without relying on external cloud services.

EcliseStore is a lightweight and efficient Java persistence framework that enables developers to store and retrieve objects directly in memory, without requiring a separate database management system. While EclipseStore offers many benefits, such as high-performance, 98% cloud database cost savings, and reduced complexity, it can be challenging to scale it to meet the demands of modern applications.
In this Workshop, you will explore how EclipseStore applications and Microservices can be scaled horizontally in the cloud with Kubernetes and what the challenges are. We will explore together how to replicate the data step by step the manual way within a docker environment and that leads us to our EclipseStore Cluster Service (MECS).
MECS enables you to build distributed EclipseStore applications by running an unlimited number of instances in parallel. You can fully automated scale-out your system on-demand and scale-in to zero on decreasing loads to save infrastructure costs. MECS is built for big-data use cases as well as for running micro clusters to scale single Microservices.