Beyond Simple Chat: Mastering LLM Interaction in Java with Function Calling, MCP, and Agents

By A N M Bazlur Rahman

Elevator Pitch

Level up your Java apps with LLMs! This session demystifies Function Calling, Model Context Protocol (MCP), and Agents. See demos showing how LLMs call Java code, manage models, and automate complex tasks. Gain practical skills to enhance workflows and build smarter applications.

Description

LLMs are reshaping the software world, introducing powerful new techniques like Function Calling, Model Context Protocol (MCP), and Agents, each unlocking unique possibilities for Java developers.

This session will simplify these powerful LLM tools by differentiating each approach. We’ll begin by illustrating Function calling via LangChain4j tools, revealing how LLM can directly invoke your existing Java methods, APIs, and databases. Next, we’ll ignite your curiosity about MCP by demonstrating how MCP aims to simplify multi-model strategies and boost interoperability. Finally, you’ll witness the exciting possibilities of Agents—intelligent systems that leverage both Function Calling and MCP to handle complex tasks and workflows with minimal intervention independently.

With practical, hands-on coding demonstrations and relatable use cases, you’ll quickly see how adopting these techniques can dramatically enhance your Java applications, streamline workflows, and deliver impressive functionality.

Whether you’re familiar with LLMs or just getting started, this session is designed to give you the confidence and skills to harness the potential of LLMs in your Java projects. By the end of this session, you’ll feel fully equipped and prepared to dive into the world of LLMs.

Notes

Outline We will walk thourgh a live coding demo for a bookstore app. (what it is, how it works, what its great for) - Tools / function calling, - MCP - Agents.

github: https://github.com/rokon12/llm-jakarta