Agentic API - the only API you will ever need
summary
Claims a single LLM plus Model Context Protocol tools can replace CRUD and many business APIs by connecting models to databases, visualization and mail services; demo connects a financial Postgres-like DB via Model Context Protocol (MCP), runs SQL, categorizes expenses across cases, builds a bar chart emailed to the user and returns the chart URL — it works end-to-end. Mentions Anthropic Claude v4, a Spring one-liner for MCP setup, and cautions about hallucinations, latency, cost and security and the need for an AI gateway.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:51AI innovations and MCP
- 3:59AI tools revolution
- 6:12Model Context Protocol
- 9:01AI agents and APIs
- 10:36API limitations
- 12:04MCP replacing APIs
- 13:54Need for AI gateway
- 15:21Agentic API architecture
- 16:23Downsides and future
- 18:31Live demo setup
- 21:02Business task demo
- 23:32Summary and Q&A
related talks
24mHow Generative AI Apps Use APIs
29mEmpowering Your Product with Agentic AI
37mLetting AI Interface with Your App with MCPs
33mThe Model Context Protocol: Connecting AI to Everything
21mIntegrate Your APIs into the New AI Marketplace
26mAPIs in Agentic world
20mLetting AI Interface with your App with MCP
26mBuilding AI-Enabled Spring Applications with Model Context Protocol (MCP)