// practical playwright training
Practise on real-world challenges: pick resilient locators, wait without sleeps, mock the network, check accessibility. Your code runs in a real browser and a hidden validator checks what happened. Start right here: an AI suggested three locators for this page, and each one is wrong.
JavaScript + PlaywrightBeginner to expertNo card needed
Remove the keyboard from the basket
getByRole('button', { name: 'Remove' })30 challenges live4 learning tracksA fresh Chromium for every runBonus points for getByRole and friends
// how it works
Each challenge is a real-world testing problem on a real practice page: delayed content, two tables with lookalike rows, a request to mock, a form to check for accessibility. Write the Playwright code in the browser editor. No setup, no imports.
await page.getByRole('button', { name: 'Sign in' }).click();Your code runs in a fresh Chromium and a hidden validator checks what actually happened on the page, not just whether your code ran. You get pass or fail, the logs and a video of the run.
After your first pass, compare your code with the reference solution. Earn points, bonus points for resilient getBy… locators and a daily streak as you move through four tracks, from Foundations to Expert & Verification.
Build confidence with roles, labels, forms and the basics of waiting, then move up through Core Interactions, Advanced Techniques and Expert & Verification.
An assistant wrote the test. Find the brittle selector and the fixed sleep, explain why they break, fix them, then prevent the next one.
Get notified about Review the AIPages that render late and change under you. Your test runs five times and must pass every time, without fixed sleeps.
Get notified about Flaky Marshlands// if you test manually today
You already know what to check. Learn to automate it one readable step at a time: open a page, click, fill a form, wait for results.
Try challenge 1, no sign-up// if you’re new to automation
Practise the habits that keep tests stable: locators by role and label, waiting without sleeps, scoping to the right row.
Start with Wait for Dynamic Content// if you already automate
Full journeys, network payloads, device emulation and accessibility assertions, ending with debugging a broken test and writing your own checks.
Start with Inspect Network Payload// who builds it
“The hardest part of automation is rarely the framework. It’s the small, repeated decisions: which locator to choose, why a test passes locally and fails in CI, when to wait and when to assert. Those skills come from practice, not documentation.”
// replay theater · coming soon
Step through a passing run one action at a time and see exactly which element each locator found.
Join the waitlist// replay theater · coming soon
Join the waitlist for early access to step-by-step browser replays.