The following summary was written by AI.
Gil Zilberfeld’s talk explains how AI can help testers work faster and cover more ground in API testing, using the CAT API as a playful example. He shows how AI can generate test plans from an API schema, suggest edge cases, organize tests by risk or category, create input and expected-response data, build setup and cleanup scripts, generate or refactor automated tests, and support exploratory testing by producing small API-call sequences. His main caution is that AI output still needs human review: testers must decide what matters, write meaningful assertions, validate generated code, and define what “good” test plans and automation look like for their team. The core message is that AI should not replace tester judgment; it should handle repetitive structure and drafting so humans can spend more time on risk, coverage, and quality decisions.
AI-assisted development can accelerate coding, but faster output does not automatically lead to better delivery. In this session, Nick Daniel and John Halberstadt examine how teams can use Agile ways of working to bring AI into the full delivery proc…
