The Role of QA When Agents Write the Tests: Why Quality Follows Context

When Kevin Goldsmith joined Microsoft, new graduates were often pushed into Software Development Engineer in Test, with the promise they could become “real” developers later. Decades later, a startup leader he mentors is asking a version of the same question about a senior manual tester on a team where agents now write much of the code and many of the tests: is this role worth what it costs?

Kevin’s answer traces how the QA role has been rebuilt over his career. It starts with Microsoft’s adversarial test org, moves through a startup whose Extreme Programming champion dismissed manual testing, the industry’s shift to offshore testing, and continues to Adobe, where testers were involved from the start of every project, and Spotify, where they were full members of the squad. The pattern he finds is about context. The value of testers, and how valuable they were seen to be, followed how much context they had and how close they sat to the work. Agents have their own kind of context, and Kevin explains why he’s come to think of an agent as the ultimate offshore tester.

He offers three quality questions for checking whether a team still has holistic quality judgment, whoever owns it: who decides what matters, who sees the whole, and who uses it like a customer. He also covers how to evaluate your existing testers, why you should staff quality to risk rather than headcount, and three ways leaders undermine quality while trying to move faster.

It’s for engineering leaders deciding what to do with their QA function as agents take on more testing, and for testers figuring out where their value lies now.

Subscribe to the Podcast

Don't miss future episodes! Subscribe on your favorite platform: