2 posts tagged with "documentation"

eConnect Core v11 ships with a new REST API, and we wanted the documentation for it to be good enough that a developer — or an AI agent — could go from nothing to a working integration in an afternoon. Here's what we built, and how.

Executive summary

• The v2 API: a REST API over all of eConnect Core — 688 endpoints across 14 modules, JSON in and out, with JWT bearer tokens for sign-in.
• A new developer section: getting started, an architecture page with diagrams, a map of the whole API, and a reference page for every module.
• Guides for pushing data in: step-by-step pages for sending casino ratings, TITO tickets, POS transactions, face detections and plate reads into eConnect from your own systems.
• A downloadable OpenAPI schema, per release, so you can generate a typed client in your own language. Version 11.0 is up now.
• Generated from the server itself: the reference pages come from the schema the server publishes, so the docs describe the API that actually runs.

eConnect Core v11 is the biggest release we have ever shipped — a brand-new client on the desktop and in the browser, Ace, Semantic Search, the Workspace and a lot more. A release like that needs a manual to match, so we wrote one from scratch.

Executive summary

• A new user manual for v11: 155 pages across 11 chapters, from your first sign-in through every module to administration. It replaces the WPF-era guides.
• Pictures on almost every page: 95 screenshots, cropped to the thing being described, with red arrows and numbered steps where your eye needs to land.
• Every screenshot comes from a live system, automatically. When the interface changes, we re-run a script instead of re-taking a hundred pictures by hand — so the manual keeps up with the product.
• Privacy is built in. Names, faces and plates are masked before the picture is taken, and a screenshot that fails a check is never published.
• Rebuilt release notes: the v11.0 release notes are now written for the person deciding whether to upgrade.