There's no AI without APIs
summary
Argument that modern AI agents require APIs to act, not models alone, and that making APIs machine-ready fixes hallucination and broken workflows. Postman defines eight characteristics (machine metadata, rich error semantics, full introspection, consistent naming, speed, discoverability), demonstrates a working pipeline with Postman Flows AI agent block calling four news APIs and posting a 3,500‑char newsletter to Slack, and promotes MCP as a USB‑C style standard; Postman Public API Network lists 100,000 APIs.
AI-written summary · about
chapters (8)
- 0:00Introduction and Welcome
- 0:37The Impact of AI on API Design
- 2:45Characteristics of AI-Ready APIs
- 10:47Postman Tools for AI-Ready APIs
- 13:00Building AI Agents with Postman Flows
- 18:33Understanding MCP and Its Importance
- 21:10Actionable Steps for AI-Ready APIs
- 22:44Conclusion and Final Thoughts
related talks
22mYou Can’t Do AI Without Quality APIs
03mHow can AI supercharge your API workflows?
26mIf your APIs aren’t agent‑ready, your AI agent is just chat: [...]
13mAre Your APIs Ready for AI Agents?
52mThe API-Powered AI Revolution: Preparing for the Future
32mYour APIs Are Not Agent-Ready: How to Test What AI Agents Actually Do
27mAI-Enhanced API Documentation [...]
21mBingo! Making Sense of AI-Ready APIs in a Buzzword World