Using MCP to Orchestrate AI Agents with OpenAI SDK, AugmentCode, and Qdrant

Multi-client protocols (MCP) are gaining momentum (with recent adoption by OpenAI and Google) as a new standard for agent interoperability.

We’ll show you how to harness the power of MCP to connect AI coding tools like AugmentCode with OpenAI’s new Agents SDK, all backed by Qdrant for fast vector search.

Speakers

You’ll learn:

• What MCP is and why it's suddenly everywhere

• The fundamentals of MCP: how it enables agent communication via a shared event bus

• Setting up an MCP server and connecting multiple clients (OpenAI Agents, AugmentCode, custom tools)

• Using Qdrant as a vector backend for memory, code context, and retrieval-augmented workflows

• Where Qdrant fits in as the vector search layer for agent memory and tool use
  
  

• Real-time agent orchestration inside your IDE, with multi-agent collaboration on tasks like code generation, refactoring, and debugging

• Real-world demos of coding agents collaborating in an IDE

• How to extend these tools for your own AI developer workflows

Whether you're experimenting with AI-supported coding or building your own developer agents, this session will give you practical, future-ready insights into using MCP with real tools.