Welcome to the 325th issue!

Looking for a new challenge?

You may want to check out QA Anywhere.

It's a free job portal created by Denis Bojkovski that frequently adds new remote job postings worldwide.

Happy hunting! 🙂

Dawid Dylowicz

Someone on Reddit shared a fresh perspective on the AI's impact on testers' roles, saying it takes away the mundane work so that we can focus on what matters. It sparked a lively discussion, with some people agreeing and others thinking otherwise.

Even though some may be Feeling exhausted due to AI, Nate Custer advises to Think Differently About AI and Testing.

AI is now doing a lot of the building, so the question of who checks the work is back. Keith Klain explains that new studies on judge bias prove the old testing mindset idea was right all along.

Similarly, Paul Grizzaffi points out that GIGO: We Still Need To Test Our Assumptions while Aaron Boadi wonders You Cannot Test Everything. So What Deserves More Attention?.

It's a good question someone asked the testing community, followed by advice from the community on what to do, including The QA Engineer Roadmap for the AI age.

Julia Pottinger shares a simple test to check if you can sign off on an AI-assisted change and where to record that sign-off in the pull request.

Also, Richard Forjoe explains why it's better to Stop Trying to Keep Up With AI Code Generation.

Rick Crawford describes six gaps that faster code delivery exposes for the testing stages. It's followed up by two more articles in the series:

• 5 QE Pipeline Metrics That Show Where Quality Is Leaking
• QE Platform Strategy: 3 Paths and Where Each Hides Its Cost

Wondering whether unit tests should be a tester's responsibility? Bas Dijkstra explains and shows three ways testers can add value, from reviewing coverage to trying mutation testing alongside the developer.

Similarly, on the practical side, Vahid Faraji asks: Mock It or Don't? How to Decide What to Mock in Unit Tests

Is AI development changing the test pyramid? Here's a new take on the possible approach called the Test Rocket, built around how AI coding agents use tests with equal unit and integration layers.

At the same time, Gil Zilberfeld looks at what coding agents need and asks: Testability Is a Feature. Does Your Code Agent Know About It?

Does your regression test suite only grow? Anton Gulin shows five checks to find which tests are worth keeping, including a simple way to catch always-green tests.

And if you're up for a more philosophical read, have a read of this thought-provoking article by Yaron Assa: The dust your test suite is kicking up is blocking the light.

Tired of fixing selectors after every front-end refactor? Surya Kulshreshtha explains a ladder of fallback strategies, from test IDs to ARIA roles, with AI healing as a last resort.

Looking for places to practice your testing knowledge? Ujjwal Kumar Singh created a few cool playground projects, with the latest one focused on hands-on API testing skills.

Sage Holloway explains how /init, skill.md files and Playwright MCP or CLI fit together and why generated selectors and credentials need review before merging.

If you're looking for ways to improve trust in your tests, Jan Molak shares Serenity/JS 3.45's new built-in HTML Reporter, with confidence scoring, flaky detection and error clustering.

Wondering how MCP 2.0 changes agent testing? Pramod Dutta covers the new stateless handshake, OAuth hardening and a simple four-layer test plan for AI agents. Also check out the MCP 2.0 Breaking Changes.

Nicola Lindgren gives a review of The Quality Coach's Handbook and shares the ideas that stood out, including why putting one person in charge of quality can let the rest of the team off the hook.

Have you heard of the Agentic Playwright repository by Ivan Davidov? It's a handy solution that scaffolds a new test framework with rules designed for AI agents. In this 35-minute recording, Alex Khvastovich tests it out.

AI: "Just rerun it..." 😂

Thanks for reading!

If you enjoy this newsletter and find it helpful in becoming a better tester, please consider sharing it with others.

Dawid Dylowicz