Arena

Building Scalable AI Systems with Spring AI and MCP

Christian Tzolov

Spring AI lead developer

Broadcom

Building Scalable AI Systems with Spring AI and MCP

The integration of Large Language Models (LLMs) into applications has evolved beyond simple API calls. Today's challenge lies in creating sophisticated AI agents that can meaningfully interact with real-world systems.

This talk walks through the journey from basic LLM integration to building fully functional AI agents using Spring AI and the Model Context Protocol (MCP). You'll learn how to choose between predictable Workflows and flexible autonomous Agents, understanding the trade-offs each approach brings.

The Spring AI and MCP Java SDK project leads will demonstrate practical patterns for building Agentic systems that can seamlessly interact with web services, file systems, and business tools. Through live coding examples, you'll see how Spring Boot starters and MCP's standardized interfaces simplify the development of AI-enabled applications.

Attendees will learn:

  • Architectural patterns for building scalable AI agents

  • The benefits of MCP's capabilities and standardized interfaces

  • Best practices for balancing simplicity with sophisticated features

  • Practical approaches to model portability and structured output handling

#Java #MCP #GenAI #Spring

Christian Tzolov

Biography

Christian Tzolov is a R&D Software Engineer at the Spring Framework team in Broadcom

Lead for the Spring AI and Spring AI MCP projects, contributor to various Spring and ASF projects including Spring Integration, Spring Cloud DataFlow.

His work focuses on system integrations, distributed data processing, data engineering and AI.