Crafting intelligent applications with Quarkus/Langchain4j

Description

This workshop introduced participants to the concepts of building Large Language Model powered applications in Java using Quarkus and the LangChain4j library. Together, they developed an enterprise application that integrated generative AI. Starting from the basics, the workshop progressively added more complex features, showcasing the different aspects of working with LLMs. Key engineers from the Quarkus and LangChain4j projects delivered the workshop, providing participants with the opportunity to ask all their questions.

By the end of the day, attendees gained a good understanding of prompt engineering, model parameters, AI security aspects, and RAG.

Topics

Basic LLM concepts
Prompt Engineering Techniques
Chat Memory: Making LLMs Stateful
Structured Data Extraction: Getting POJOs from LLMs
Easy and Advanced RAG (Retrieval-Augmented Generation): enhancing LLMs with custom data
Agents and Tools: enhancing LLMs with custom skills
Enterprise Features: Observability, Fault Tolerance, Content Moderation
Evaluating Non-Deterministic Applications: Are We Improving?
Data privacy consideration for AI in enterprise applications
How to future-proof your developer career

Requirements

The workshop assumes familiarity with Java development and tools, however, no prior AI knowledge is required.
Check this document for instructions what you have to prepare.

Speaker

Dimitris Andreadis

Engineering Director at Red Hat, Quarkus Team, Cloud Native Runtimes

Lize Raes

Team Lead & Software Engineer in Bioinformatics and Generative AI

Dmytro Liubarskyi

Creator and Lead developer, LangChain4j