Building Agentic AI systems: A2A (Agent2Agent) & MCP Protocols & Serverless

Exciting technical session where Dan Dobrin will decode the latest protocols powering intelligent Agentic AI systems. We will explore how Model Context Protocol (MCP) and Agent2Agent (A2A) Protocol are transforming the development of enterprise-grade AI applications in serverless environments. While many RAG implementations require Python, Dan will demonstrate how these powerful protocols can be implemented in enterprise Java environments. This session offers valuable insights for developers across all major cloud platforms, as these emerging open standards are designed for multi-cloud, interconnected systems.

What You'll Learn

• - How Model Context Protocol (MCP) enables robust tool usage and integration in LLM applications
• - The architecture and implementation patterns of Agent2Agent (A2A) Protocol for creating multi-agent, collaborative AI systems
• - Practical approaches to building intelligent agentic systems in serverless environments
• - Enterprise-grade implementation strategies beyond proof-of-concept demos

Whether you're working with AWS, Azure, or Google Cloud, these open protocols represent the future of intelligent system development. Join us to explore how these technologies can enhance your AI applications with more sophisticated reasoning, planning, and collaboration capabilities.

Agenda
6:00 - 6:10 PM: Welcome and Introduction
6:10 - 6:50 PM: Main Presentation:

• Agents vs Workflows: Understanding the evolution from simple to complex systems
• Agentic Capabilities: Function calling, planning, and reflection
• RAG implementation in enterprise systems
• Model Context Protocol (MCP): Architecture and implementation
• Agent2Agent (A2A) Protocol: Extending MCP for agent collaboration
• Java implementation patterns for enterprise environments
• Deployment in serverless environments (Cloud Run)

6:50 - 7:20 PM: Demo and Q&A
7:20 - 7:30 PM: Wrap-up and Manning Publications Raffle

About the Speaker
Dan Dobrin is an Enterprise Application Architect at Google Cloud with extensive experience in enterprise Java applications and cloud technologies. He recently presented multiple sessions at Google Cloud Next '25, including talks on GenAI, LangChain4j, Gemini Code Assist, and Agentic RAG systems. Dan previously spoke at Serverless Toronto about High-Performance Java on Google Cloud Platform, and is a frequent speaker at Spring I/O conferences. With a focus on practical, production-ready implementations, Dan brings real-world insights into cutting-edge technologies.

Who Should Attend

• Software architects and developers interested in building intelligent AI systems
• Enterprise Java developers looking to implement advanced AI capabilities
• Cloud practitioners working with serverless technologies
• Technical leaders evaluating emerging standards for AI agent development
• Anyone interested in the future of LLM-powered applications in production environments

Format
This is a virtual event. Connection details will be shared with registered attendees.

Don't miss this opportunity to learn about the protocols shaping the future of intelligent systems development from an experienced enterprise architect. Whether you're just getting started with AI or building sophisticated agent-based applications, this session will provide valuable insights applicable across cloud platforms.