Documentation
The full syllabus, in the open.
This is the actual course. Every chapter includes entry checks, a common-mistakes box, an exercise and a quiz.
Foundations
- Zero to automation testing — manual vs automated, setup from scratch
- AI for testers — prompting AI tools for test cases, test data and bug reports
- What is Playwright — browsers, locators, auto-waiting
- JS + TS essentials — only the ideas the framework uses
- BDD with Gherkin — Given/When/Then, tags, flow
Framework craft
- Page Object Model — the three rules, BasePage, fixtures
- Framework tour — every folder and file, end to end
- API testing in the framework — request contexts, REST checks, API-driven test data
- Handling hard cases — iframes, shadow DOM, sessions, Allure
- AI-assisted framework coding — generate, refactor and review page objects and tests with AI
- Solved capstone — empty folder to green suite
Professional edge
- Advanced concepts — sharding, traces, CI, debugging playbook
- Playwright MCP + AI agents — agent loop, healer, triage
- AI in CI/CD — AI failure analysis, self-healing locators, Jenkins and GitHub Actions
- Interview bank — Q&A, drill plan, scoring rubric
- Architect level — strategy, metrics, flaky-test policy
How our AI test agents are built
Three layers. One hard rule.
1
Intent (AI)
Requirements + RAG context → an LLM plans a structured test intent (JSON).
2
Validation
Schema checks, selector verification against the DOM and safety guards.
3
Execution
An MCP server + Playwright run the validated intent and capture evidence.
!
The rule
No LLM calls during execution — same intent, same actions, CI-safe results.