QA / Automation Engineer Job Simulation
Meridian's test suite has been green for six weeks - including the build that broke cancellations. Find the tests that are lying, then design the strategy that replaces re-run-until-green with a signal engineers actually trust.
Join 100,000+ One Roadmap-certified candidates
Nina WalshQA Lead, Meridian
Welcome to Meridian. I'm Nina, QA Lead. We run bookings for clinics and studios, and last week we shipped a bug that stopped customers cancelling - through a completely green build. Your first ticket is to tell me which of our tests are lying and why. Your second is to design what our suite should be, given we deploy twice a day on a fixed CI budget. Fair warning: our engineering lead wants to solve this by adding more end-to-end tests, and I need you to take that argument on rather than talk around it.
Your manager for this simulation. Reviews feel this real too.
Why finish this simulation
A risk-free way to experience the QA job before you're hired - audit a real suite that gave false confidence, design a strategy under a real CI budget, and back your applications with evidence rather than a list of tools.
The failure mode nobody teaches. A red suite is annoying; a green suite on a broken build costs customers. This simulation is built around the second one.
Flaky and worthless are different. Most candidates treat every failing test as flaky. Telling the two apart is the judgement QA leads actually hire for.
Automation judgement, not tool trivia. Nobody asks what Playwright's API is. They ask why your test passed when the feature was broken - and this is that conversation.
Strategy under a real budget. Two deploys a day, 34 minutes of E2E and no new runners. Designing inside constraints is the whole job.
An argument you have to win. QA-102 has an opposing position built in. Engaging it is scored; asserting past it is not.
Feedback that names the gap. Every submission returns per-question results and a rubric read of your notes, so a retry is targeted rather than a guess.
How it works
01Pick up a ticket
Real tasks land on your board like a Jira queue. Download the dataset and work in your tool of choice.
02Submit for review
Your work enters review - results within 24 hours (usually much sooner). The clock pauses while you wait, like a real take-home.
03Pass, then progress
Pass to unlock the next ticket. Clear every ticket to earn a certificate and a spot on the weekly leaderboard.
Skills you will learn and practice
- Telling a flaky test apart from a worthless one
- Spotting a test that asserts its own mock
- Why fixed sleeps pass locally and fail in CI
- Finding order-dependent tests hiding in shared state
- Asserting user-visible behaviour over CSS classes
- Quarantine policy instead of retry-until-green
- Shaping a suite to a real CI budget
- Measuring suite trust by signal-to-noise, not coverage
The tickets
What you'll learn
- Why a test that mocks the thing it tests passes on a broken build
- How fixed sleeps turn a suite green locally and red on shared CI runners
- How shared state makes a passing suite depend on execution order
What you'll do
- Audit a real Playwright file against 40 runs of CI history
- Separate the flaky tests from the worthless ones
- Write the review that changes a team's re-run-until-green habit
Weekly leaderboard
Highest ranking points top the board - your score counts 80%, your speed 20%.
Your Future Certificate
Preview only. Clear both tickets to earn your personalized certificate - free.
Verified Certificate
Unique verification URL & QR code
LinkedIn Integration
Add to your profile with one click
Performance Analysis
Detailed insights into your results
Issued by DPIIT-recognized & MSME-registered company
Clear it, prove it
Pass both tickets and your verifiable certificate is free - add it to LinkedIn right away. Want to go deeper? Generate a personalized AI review of your performance for a one-time ₹99.