Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
MCP: Standardizing AI Connections
This talk explains the Model Context Protocol (MCP), demonstrating how it standardizes AI-tool connections, simplifies integrations, and enables quick deployment of AI applications.
In this video, Santiago and Alden break down the Model Context Protocol (MCP) - Anthropic’s standard for connecting AI applications to external tools and data sources.
Santiago explains the fundamentals:
What MCP is and why Anthropic created it (hint: “models are only as good as the context provided to them”)
The USB port analogy - how MCP creates a universal interface for AI connections
The dramatic simplification from n×n custom integrations to just n+n standardized connections
Alden demonstrates real-world implementation:
Live demo showing how to create a RAG system in minutes using Custom GPT
Step-by-step walkthrough of ingesting a website (Vanguard institutional) into a searchable knowledge base
One-click deployment that generates an MCP-compatible endpoint
Practical example of integrating with OpenAI’s API in just a few lines of code
Perfect for developers and business users who want to understand how MCP can multiply their AI capabilities by seamlessly connecting their knowledge and data to any MCP-compatible application - from Claude Desktop to Cursor to custom AI agents.
Key takeaway: What used to require complex custom integrations for each tool can now be done with a single standardized connection, opening up an “n by m connection multiplier” for your business data.
Hosted MCP Servers provide managed, production-grade RAG endpoints for AI agent data access.