How do you apply test-first to an artificial intelligence application? That was a question posed to me during one of my ATDD/BDD workshops. Here are my thoughts. I got inspiration from recent articles in IEEE Spectrum and Communications of the AC. I welcome additional suggestions from you. The examples here represent some of the areas in which AI is currently used. Another article will explore acceptance testing for autonomous vehicles.
“How Do We Know We’re Done?” is a question often asked by agile teams. I’d like to show a narrative of how this question might be answered. As the example, I’ll use one from James Shore’s new edition of Art of Agile Development, Chapter 9, Customer Examples. He describes a meeting between the product owner and the developers to gather information on a story The section ends with the developer’s taking a picture of a whiteboard of a business rule and heading off to develop. Here’s a possible narrative for the rest of the story of “How Do We Know We’re Done”.