The latest articles on DEV Community by Andreas Lottes (@andreas_lottes_920b928bc9). https://dev.to/andreas_lottes_920b928bc9 https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4028371%2F61382c57-c1f7-4ee8-9d02-0fb578441236.jpeg https://dev.to/andreas_lottes_920b928bc9 en Andreas Lottes Tue, 14 Jul 2026 08:25:12 +0000 https://dev.to/andreas_lottes_920b928bc9/why-i-gave-claude-code-a-qa-tester-that-actually-opens-a-browser-56fi https://dev.to/andreas_lottes_920b928bc9/why-i-gave-claude-code-a-qa-tester-that-actually-opens-a-browser-56fi <p>Claude Code can build a feature in minutes. The problem isn't speed — it's that "finished" usually means the code compiled and the tests it wrote itself went green. Nobody challenged the idea, nobody reviewed the diff with a critical eye, and nobody actually used the thing.</p> <p>I wanted the opposite of a yes-man. So I built aSPARK: an AI product team for Claude Code that runs a gated delivery loop, where each step has to earn its way to the next one. This post is about the one decision that changed how much I trust the output — making QA click through the real app in a real browser — and what enforcing quality gates on an AI coding agent actually feels like in practice.</p> ai claude showdev testing