Each one is a public repository with its own README. Most run in CI on GitHub Actions.
UI test framework for saucedemo.com: Python, Selenium and pytest with a Page Object Model, run in CI on GitHub Actions.
17 tests across login, inventory, cart and checkout, tagged smoke or regression. Failures save a screenshot, and every CI run uploads an HTML report.
AI-assisted test generation with Claude Code: user stories to test cases to Playwright tests, with every AI mistake logged.
Nine user stories became 31 reviewed test cases and 31 Playwright tests. A person reviews each step, and the AI's mistakes are written down in a review log.
Automated quality checks for an AI assistant's answers: grounding, citations, refusals, consistency and golden questions, with a CI quality gate (pytest, GitHub Actions).
Three deliberately broken assistants must each fail the checks, and the tests caught a real ranking defect while I built it.
Property-based tests (Hypothesis + pytest) for a wage compliance rules engine, with a rule traceability matrix and a CI quality gate on GitHub Actions.
I broke the engine three ways on purpose and the tests caught each one. A CI gate fails if any rule has no tests.
-
API tests for the Restful-Booker API in pytest and in a Postman collection run with Newman, with GitHub Actions CI.
-
Cypress end-to-end tests for the-internet.herokuapp.com, covering forms, dialogs, dynamic content and HTTP status codes, with GitHub Actions CI.
-
Manual QA test plan, 42 executed test cases (31 pass, 11 fail) and 11 bug reports for the SauceDemo store.
-
Apache JMeter performance tests for a public REST API with smoke, load and stress scenarios, pass/fail thresholds and GitHub Actions CI.
-
Cross-browser Selenium and pytest suite that runs in parallel on a Dockerized Selenium Grid (Chrome and Firefox), with GitHub Actions CI.
-
BDD test suite in Gherkin with Behave and Selenium (Python), using the Page Object Model and GitHub Actions CI.
-
SQL data quality checks run with Python and pytest: clean data passes every rule and injected defects are caught.
-
Automated WCAG 2.1 accessibility testing with axe-core, Selenium and pytest, with a summary report and GitHub Actions CI.
-
API contract tests for a public REST API using JSON Schema, pytest and requests, with offline strictness tests and GitHub Actions CI.
-
Visual regression tests: Selenium + pytest + Pillow screenshot comparison with baselines and red diff images, CI on GitHub Actions
-
Python log analysis tool that finds error spikes after a deploy and traces which service failed first, with pytest tests and a CI regression gate (synthetic logs)
-
pytest framework that grades an AI model's answers with code checks (accuracy, format, tool use, hallucination, safety) and a CI quality gate; offline mock model, optional Anthropic API
-
Flaky test detector for pytest: reruns a suite in normal and shuffled order, finds flaky and order-dependent tests, and fails a CI quality gate on any test not in quarantine (Python, pytest, GitHub Actions)
-
Security tests for a demo shop API in pytest and Flask, mapped to the OWASP API Top 10: ten weaknesses can be planted one at a time, and the suite must catch each one, only through its own tests (Python, pytest, Flask, GitHub Actions)
-
Java UI tests with Selenium and JUnit 5 for a practice shop, using a Page Object Model: eight weaknesses can be planted one at a time, and the suite must catch each one (Java, Maven, GitHub Actions)
-
Small Apache JMeter smoke test against the public JSONPlaceholder API, with status and JSON assertions and a GitHub Actions workflow that fails on any failed request.