Question 1

Your team has 900 UI end-to-end tests and 100 unit tests. The suite takes 3 hours and fails randomly. What is the main structural problem?

Accepted Answer

The test pyramid is inverted. Most coverage sits at the slowest, least stable layer. The testing pyramid says: many unit tests, some integration/API tests, few UI tests. When the pyramid is inverted, the suite becomes slow and unstable no matter which tool you use.

Question 2

A test passes locally but fails in CI about once in five runs. What should you do first?

Accepted Answer

Run that single test many times in a loop to reproduce the failure. Debugging flakiness starts with reliable reproduction, then finding the wrong timing assumption. Sleeps, skips and retries all avoid the real cause.

Question 3

What is the main benefit of the Page Object Model (POM)?

Accepted Answer

UI changes require fixes in one place instead of in every test. POM is about maintainability: locators and interactions live in one reusable class per page, so tests read like scenarios and UI changes have one fix location.

Question 4

Which check belongs in an API test rather than a UI test?

Accepted Answer

That the server rejects a request with an expired token with status 401. Push everything that is server logic (validation, auth, data rules) down to API tests. Keep UI tests for what users actually see and do in the interface.

Question 5

Why do teams run automated tests in a CI pipeline on every pull request?

Accepted Answer

To catch regressions before the change is merged, while the cause is obvious. The value of CI is fast feedback tied to a small change. The smaller the change that broke the build, the cheaper the fix.

Question 6

A test clicks 'Save', then sleep(2000), then asserts the toast text. What is the professional improvement?

Accepted Answer

Replace the sleep with an assertion that waits for the toast to appear. Wait for what you expect, never for an amount of time. Modern assertions retry automatically; in Selenium use explicit waits.

SDET Interview Preparation

Your team has 900 UI end-to-end tests and 100 unit tests. The suite takes 3 hours and fails randomly. What is the main structural problem?

Design a login test suite

Stabilize a flaky flow