Fullstack Engineer - SDET Depth (Quality Engineer)

You will own the quality net for a client-delivery squad: the system that makes a missing spec, a dropped requirement, or a silent regression impossible to ship unnoticed. You are a fullstack engineer who builds the fastest and hardens everything, and you spend that speed engineering quality in rather than inspecting it at the end.

What you will do:

Build and own the artifact net: a Definition of Ready that stops a story entering build without its acceptance criteria and spec, automated CI checks that fail when a change has no test or breaks a covered path, and a Definition of Done enforced at release.

Own the regression system: test architecture, CI gates, the automated regression and smoke suites, defect management, and release QA.

Build your own modules end to end (API, UI, data, tests), like every engineer on the squad.

Hold stop-the-line authority at release: block a release when the net goes red, and own that call.

Make and defend severity and risk decisions, and surface systemic patterns, not just individual bugs.

Use AI heavily to move fast, and apply the judgment that AI cannot: what to test, what to gate, what a passing test actually proves.

How you will be measured:

quality is a property of the system you build, so that it holds even when you are not in the room, and scales to the next squad without you. You succeed when the team ships predictably and the net, not your vigilance, is what catches the problems.

Must have:

4 to 7 years building production web software, with a track record of owning quality, not only writing features.

Strong test automation: can design a test architecture and CI gates, not just write individual test cases (Playwright, Jest/Vitest, or equivalent).

CI/CD experience: has built or maintained pipelines that block bad changes (GitHub Actions or equivalent).

Sound defect judgment: can classify severity (blocker vs data-integrity vs minor vs cosmetic) and defend the call. Treats a bug that shows the right thing on screen but stores the wrong value underneath as serious.

Uses AI as leverage and verifies it: generates with AI, then curates and catches where AI-generated “passing” tests prove nothing.

Communicates and holds a line: can tell a team a release is not ready, with evidence, under deadline pressure.

Nice to have:

Experience defining and enforcing a Definition of Ready / Definition of Done as tooling, not a wiki page.

Requirements traceability or test-coverage-gap analysis experience.

B2B / client-delivery or consulting-software background (multiple parallel projects).

Exposure to AI-feature delivery (evaluation sets, ground truth, regression on prompt changes).

Not required (and not what we are hiring for):

Manual-only QA background with no automation or coding.

“Tester at the end of the line.” This seat builds quality in from the first sprint and owns its own modules too.