The Full Agentic QA Loop for Web Applications — Author, Evidence, Assure, Maintain

Most "AI testing" demos stop at generating a test. Real QA is a loop: you author coverage, capture what actually happened, assure it against intent, and keep it all from rotting as the product changes. In this hands-on session we run that full loop end-to-end with kane-cli, an agentic test tool that drives a real browser and authors runnable tests from plain-language requirements.

You'll leave with a working pipeline on your own machine. We cover:

Author — Turn a one-line requirement into concrete scenarios and cases with kane-cli generate, then let the agent execute them against a live browser instead of hand-writing selectors and waits.

Evidence — Capture every run as a portable, framework-agnostic .evidence pack: the DOM, actions, network, and assertions that prove what the run did — not just a green check.

Assure — Move from brittle exact-match assertions to intent-level verification, so a test fails when behavior breaks, not when a class name changes.

Maintain — Close the loop: keep tests and their knowledge base accurate as the app evolves, using agentic review to catch drift before it ships.

By the end you'll understand where AI genuinely replaces manual QA toil, where a human still owns the gate, and how to wire the author-evidence-assure-maintain loop into your own stack. Bring a laptop with Node installed; we build and run everything live.