NEWS
The Death of Determinism: How AI Forces Us to Rethink Testing
Padget Avery explains how testing AI differs from the traditional approach, requiring moving from strict assertions to thresholds, metrics, and human review in pipelines.
Moreover, Katja Obring shares thoughts on What I don't understand about AI evals (yet) and how The AI evals field chose a flawed tool and stuck with it.
medium.com
We Need to Stop Queueing Quality
Should teams keep a separate testing stage on their boards? Jitesh Gosai shares a simple analogy to show how it can delay feedback.
Similarly, Joshua Bihun wrote good articles on What Acceptance Criteria Actually Ought to Say and how to Shift-Left Quality From Day One.
substack.com
When building is cheap, quality becomes a bigger differentiator for your software product
AI speeds up coding, but at what cost? Esben Bager raises a good point β software advantage will now be driven primarily by quality and reliability rather than the number of features.
Similarly, Josh Grant points out that Software Quality is Heading Toward a Revolution and, in a similar context, Maaike Brinkhof says to Stop selling gold when all you've got to offer is a process.
medium.com
Why I am not worried about AI replacing me
With the rapid, constant development of AI, you may wonder whether it poses a threat to the tester's role. Matthew Sullivan doesn't think so, for a few reasons.
Similarly, Brad DerManouelian explains why AI Isn't Killing Manual QA. It's Supercharging It.
medium.com
AUTOMATION
AI and Testing: Using Local Models for Testing
Jeff Nyman continues a detailed series of articles on AI in testing, and shows how to set up and run a model on your machine to create automated tests with evaluation metrics. You can also read a follow-up article about Using Model Pipelines for Testing.
testerstories.com
My Engineering Manager asked for a test coverage dashboard. Here's what I built
Who wasn't in that situation? It's interesting to see how Osanda Deshan Nimalarathna approached building a dashboard for tracking automated test coverage.
medium.com
Only 41% of Testing Is Automated in 2026. More AI Tools Won't Fix That.
Pramod Dutta shares insights from the recent Leapwork survey on the state of test automation, revealing some (un)surprising data.
medium.com
Rethinking performance testing for the modern web
An insightful article from Shyamal Raju explaining why browser performance is important for user experience and demonstrating how to include browser flows in load testing.
On this note, Oleh Koren shares good thoughts on Why Percentiles Matter More Than Average Response Time in Performance Testing.
medium.com
Test Plan Driven Development (TPDD): A Developer-Friendly Alternative to TDD
This sounds novel. Vigor Zang recognises the friction with practising Test-Driven Development and suggests a different approach to ensure quality since the early stages of the software development life cycle.
Moreover, Sefa Esendemir wrote about more levels of the test pyramid β From Writing Code to Breaking It.
medium.com
TOOLS
17 Playwright Testing Mistakes You Should Avoid
There is some great, practical advice on how to improve your Playwright tests to run more smoothly and fail less often. By Yevhen Laichenkov.
github.io
Deep Dive into Playwright CLI: Token Efficient Browser Automation
Two weeks ago, I mentioned the release of Playwright CLI. Now we can learn more about it, thanks to this detailed overview by Pratik Patel.
testdino.com
k6-cucumber-steps: Run K6 Performance Tests with Cucumber BDD Syntax
Cucumber for performance tests? Chetachi Enyimiri thought "Why not!" and demonstrated how to achieve that with k6.
Also, Marlo Henrique advises Using SharedArray with k6.
medium.com
Making Code Coverage Work with Playwright
Sayed Afnan Khazi tells us how they've been looking for a way to get Playwright test coverage in a non-trivial project setup and how the playwright-coverage library made it happen.
Speaking of that, Bhavik Gandhi wrote a helpful guide on How to Intercept and Modify Network Responses with Playwright.
medium.com
Migrating From Cypress.env To cy.env and Cypress.expose Methods
There's a major change coming in Cypress v16 and Gleb Bahmutov shows how to prepare for handling the environment variables and secrets in a new way.
Also, Jennifer Shehane has just announced another improvement that allows you to Fix Cypress CI failures caused by no spec files found.
glebbahmutov.com
AND...
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
Welcome to the 305th issue!
Skills are the new way for instructing AI agents to generate better results.
And now we have it for testing, too!
Happy (skills) testing! π
Dawid Dylowicz