Hi everyone,
This is the sixth video of the series "Breaking down Agentic Architecture". In this video, we talk about the Model Context Protocol, or MCP.
We start by analysing the need of this protocol, going back to the initial tool calling in LLM models or assistants. Then we introduce the architecture of the MCP and its capabilities, such as prompts, resources and tools. And we finish by giving some use case examples that can be leverage by using an MCP server.
Follow me into this topic where it can help you look into your business use cases with another set of glasses.
Let me know also your ideas around this topic :)
Cheers!
Today's chapters:
00:00 - Introduction
01:04 - Initial Tool Calling in AI Assistants
04:54 - Model Context Protocol
11:30 - Use Cases for MCP Servers
コメント