Why pgEdge thinks MCP (not an API) is the right way for AI agents to talk to databases

Here's a podcast episode description: **What if the way AI agents talk to databases has been wrong all along?** Most teams default to APIs when connecting AI to their data layer — but pgEdge is making the case that MCP (Model Context Protocol) is a fundamentally better fit for how agents actually think and work. In this episode, we dig into why the protocol layer matters more than you might expect, and what it means for the future of AI-native applications. --- Want me to try a different tone — more technical, more casual, or punchy/provocative?