Julia Muiruri

Talk

Building a Burger Ordering AI Agent with LangChain.js and MCP

This talk demonstrates how to build a simple AI agent that interacts with customers to take their burger orders, covering the design, implementation, deployment and monitoring (Observability) of the agent.

Status Available Duration 60 minutes Audience Developers, Builders

In this talk, I first engage the audience in an interactive demo of the Burger Ordering Agent, then go through the underlying architecture and design principles across the Business Layer, API Layer and the Agent Layer, and dive into the implementation details of the agent, how to use Langchain to build & customize the agent’s capabilities and observability features on Microsoft Foundry (Azure).

What the session covers

  • Designing practical Agentic systems
  • The 3 layers of THIS agent architecture: Business Layer, API Layer and Agent Layer
  • Building the agent with Langchain and Microsoft Foundry
  • Observability in the agentic world

The goal of this talk is to provide developers with a practical example of how to build an AI agent, and to share insights and best practices for designing and implementing agentic systems that can interact with customers in a natural and effective way.