See all blog posts
Build AI-infused Java applications with the Liberty LangChain4j Workshop
Generative AI is rapidly becoming part of enterprise Java development, and developers need practical guidance on building AI-powered applications that are reliable, secure, and production-ready. To help you get started, we’re excited to announce the new Liberty LangChain4j Workshop.
The workshop is a hands-on tutorial that guides you through building AI-infused applications with Open Liberty and LangChain4j. Starting from a simple chatbot, you’ll progressively add more advanced capabilities while learning the patterns and techniques used to integrate large language models (LLMs) into enterprise Java applications.
What you’ll learn
The workshop begins by introducing the fundamentals of LangChain4j and shows how to connect an Open Liberty application to an LLM. From there, you’ll explore many of the building blocks that modern AI applications rely on, including:
-
Prompt engineering and system messages
-
Streaming responses
-
Retrieval-Augmented Generation (RAG)
-
Function calling and tools
-
Model Context Protocol (MCP)
-
Guardrails for safer AI interactions
-
Observability and fault tolerance for AI workloads
After covering the core AI application patterns, the workshop moves on to agentic systems. You’ll learn how to build AI agents and compose them into increasingly sophisticated workflows, including multi-agent orchestration, supervisor patterns, human-in-the-loop interactions, and multimodal agents.
Throughout the workshop, each step builds on the previous one, allowing you to understand not only how to use these capabilities, but also why they’re useful and when to apply them.
Get started
Whether you’re experimenting with AI for the first time or looking to expand your enterprise Java applications with intelligent capabilities, the Liberty LangChain4j Workshop provides a practical, hands-on introduction to modern AI development with Open Liberty.
The workshop is available now on GitHub Pages: Liberty LangChain4j Workshop
We welcome your feedback and contributions. If you have suggestions for improvements or would like to contribute additional content, please open an issue or pull request in the workshop repository.
Acknowledgement
This workshop builds upon content from the Quarkus LangChain4j Workshop. We gratefully acknowledge the contributors to the Quarkus LangChain4j Workshop for their excellent work and for permitting us to adapt and extend it.
