MCP Explained – Build Your First MCP Server

Large Language Models are evolving in how they interact with external data, moving beyond static prompts to more dynamic, real-time integrations . Traditional approaches relying on hard-coded APIs often limit flexibility and scalability in real-world applications.

In this session, we’ll explore the Model Context Protocol (MCP) — an open standard that enables LLMs to dynamically discover and interact with data sources and tools as if they were part of their own memory. You’ll understand how MCP introduces a standardized “handshake” between models and external systems, enabling more flexible and scalable AI architectures.

The session includes a live demo of an MCP-powered LLM assistant for e-commerce , where the model actively queries product data and generates intelligent recommendations. By the end, you’ll gain a clear understanding of MCP architecture, its advantages over traditional APIs, and how to implement it in real-world systems .

This is a technical deep-dive session designed to help practitioners build more connected, context-aware AI systems .

Key Takeaways:

• Understanding MCP – how Model Context Protocol enables dynamic interaction between LLMs and external systems
• Beyond APIs – key differences between MCP and traditional integration approaches
• Real-Time AI Systems – enabling LLMs to query and reason over structured datasets
• Practical Demo – building and connecting an MCP server for an e-commerce use case
• System Design Perspective – creating scalable, flexible AI architectures using MCP

Geethanjali is an ML leader with over a decade of experience across retail and fintech, having worked with organizations like Target, PayPal, and Verizon, where she currently leads Fraud Analytics.

She specializes in translating complex data into actionable insights and is passionate about driving measurable business impact through data-driven decision-making.