Unlocking business value using multi-agent orchestration and MCP servers

Summary:
Enterprises are rapidly evolving from AI experimentation to strategic adoption, integrating AI to drive growth, efficiency, and innovation.

At the heart of these implementations lie AI agents—autonomous entities that execute business functions and collaborate with tools and other agents to deliver intelligent solutions. In this session we will deep dive into the core components of AI agents, their interfaces, and the orchestration methods that enable seamless multi-agent collaboration.

Through hands-on development, we’ll explore various orchestration patterns and demonstrate how to effectively leverage tools exposed by MCP servers to build robust, scalable agent systems.

Agenda:
Part-1: Session - Deep dive into Agents and its components

• Agents
• Tools
• Chat completions API
• Tool calling LLMs
• Agent orchestration patterns

Part-2: Multiagent orchestration hands-on

1. Supervisor - worker agent pattern based scenario using Langgraph
2. Chain of thought reasoning agent pattern based scenario.

Part-3: Leveraging MCP servers in AI Agents - Hands-on

1. Create a developer productivity agent that allows developers to perform git related tasks using agent configured with git based MCP server. (edited)