Capstone project for Test Automation and Quality Engineering
Automate the checkout test suite
Mangrove's checkout has 200 manual test cases that take two days per release, so fixes wait for the next release window and festival changes ship late. Replace most of them with an automated suite that runs in under ten minutes on every pull request: a written strategy that maps checkout risks to levels, unit tests for the price rules, API tests for the order service with its own test data, a few Playwright journeys, a contract test between the app and the order service, a load test for a 4x festival rush, and a CI pipeline with reports. Use the course's sample checkout or any small app of your own with a cart and an orders API. Plan for 8 to 12 hours.
What you hand in
- A one-page test strategy: risks mapped to levels, planned test counts and minutes per level, and what stays manual
- A repository with unit, API, end-to-end and contract tests in TypeScript
- A CI pipeline (GitHub Actions) with parallel jobs, an HTML report and traces kept on failure
- A k6 load test with thresholds and a short results note
- A README with how to run everything and the flaky-test policy
Check your work
Tick each item when your project meets it. Be strict: this is the rubric the reviewer uses.
- Most checks sit at the fastest level that can catch the bug: the strategy maps each risk to a level and the test counts match it
- End-to-end tests cover only a few critical journeys, use role, label or test id locators, set up data through the API or fixtures, and never use fixed waits
- Unit and API tests are readable and independent: behaviour names, table-driven boundaries, builders for test data, a real database for API tests, no shared mutable state
- A contract test catches a breaking change in the order response, and the pipeline would block a provider release that breaks the app
- The load test models the festival rush with an arrival rate from stated assumptions, sets thresholds on p95 latency and errors, and the note reads percentiles, not averages
- The whole suite runs in under ten minutes in CI, with timings shown, sharding or parallel workers, and reports and traces uploaded as artifacts
- The flaky-test policy names how flakes are detected, quarantined with an owner and fixed, and AI-drafted tests are reviewed against a checklist
0 of 7 checked
Model answer
The model answer opens once your capstone passes the review, so you can compare your work with an outline of a strong submission.
Hand in your capstone
Your links are checked automatically, then the work is scored against the rubric above, item by item, with comments. A pass counts toward the course certificate and shows on your profile.
Loading…