Building a Multi-Agent System with the Claude SDK (MCP + Tool Integration)
Building a Multi-Agent System with the Claude SDK (MCP + Tool Integration)
Hosted by Spritle Software | Claude Certified Partner
What happens when you move beyond a single AI agent and build a system where multiple agents work together?
Join Spritle Software for a practical deep dive into building multi-agent systems with the Claude SDK, exploring how orchestrators, subagents, MCP servers, and custom tools come together to create capable AI-powered workflows.
In this session, we’ll move from architecture and core concepts to implementation, integrations, and a live demonstration of a working multi-agent system.
What We’ll Cover
1. Multi-Agent Architecture Overview
Understand the role of orchestrators and subagents, when to split work across multiple agents, and how to design effective agent workflows.
2. Building Agents with the Claude SDK
Walk through the core setup and message loop needed to build agents using the Claude SDK.
3. Integrating MCP Servers
Learn how to connect MCP servers to your agents and extend their capabilities with external tools and resources.
4. Adding Custom Tools
Explore tool definitions, tool calls, and how agents can interact with custom capabilities to complete tasks.
5. Live Demo
See a multi-agent system in action, bringing together orchestration, MCP, and tool integration in a practical workflow.
6. Q&A
Bring your questions and get practical insights into designing and building multi-agent systems.
Who Should Attend?
This session is ideal for:
Software developers and engineers
AI/ML practitioners
Engineering leaders
Developers experimenting with Claude SDK
Teams building agentic AI workflows
Anyone interested in multi-agent architectures and MCP
Whether you're exploring multi-agent systems for the first time or already building with Claude, this session will give you practical concepts and implementation insights to take your AI workflows further.
Hosted by Spritle Software — Claude Certified Partner.
Come learn how to move from individual AI agents to connected, tool-enabled multi-agent systems.