Listed as a Master Workshop on the API World 2026 workshops page — Workshop Stage B, Santa Clara Convention Center. Included with PREMIUM and WORKSHOP passes.
In 3 hours, you’ll go from an idea to a running, agent-tested API. Requirements, spec, code, deployment, and proof it works with an AI agent. You build it, you test it, you ship it.
Your APIs were built for developers who read docs, ask questions, and figure out edge cases. Agents don’t do any of that. They need to discover, call, and chain your APIs without help. When they can’t, they move on. To a competitor. To scraping. To telling the user it couldn’t be done.
So instead of talking about what agent-ready means, we’re going to build it. From scratch. And prove it works.
Here’s what you’ll build:
- What agents actually want from your API, and the one thing they don’t
- The design decisions that separate agent-friendly APIs from the rest
- Write real requirements using markdown and API standards (we provide them, or bring your own)
- Design your API and generate an OpenAPI spec with AI
- Mock it and put it in front of Claude Code. Does the agent understand your spec?
- Build the real thing with Replit. Running endpoints, live
- Put your running API in front of an AI agent. Does it actually work?
- Your agent-readiness checklist and roadmap for your team
- Q&A and where to go from here
You’ll walk out with:
- Requirements you wrote
- An OpenAPI spec you generated
- A running API you built and deployed
- Proof it works with an AI agent
- A checklist and roadmap to take back to your team
Fair warning: this isn’t a lecture. You’ll be building for 3 hours straight.