SDET-1 (Manual + Automation QA) (HighAdvocacy)
Saleshandy
Core Focus: Test automation and quality engineering across the stack moving testing left so bugs get caught automatically on every PR, not by manual QA in production.
Tech Stack: Playwright, TypeScript / JavaScript, Cucumber (BDD), Page Object Model, GitHub Actions (CI/CD), Postman (REST API testing), MySQL / SQL, Git
Bonus: JMeter (performance), OWASP Top 10 (security), AI testing tools (Copilot / ChatGPT for test generation, self-healing locators)
About High Advocacy
High Advocacy turns a B2B SaaS company's silent happy customers into its loudest marketing channel. When a customer hits a milestone, closes a big deal on the back of your product, or drops a 9/10 NPS, we make it effortless for them to share that win publicly as a review, a social post, or a text or video testimonial verify it actually happened, and get rewarded on the spot.
Teams run all of it from one workflow: AI-assisted drafts written for the customer, a single clean approval queue, email and Slack notifications, and a Proof Library that stores every approved asset to reuse in a sales deck, a landing page, or a comparison page in seconds. It's an early, fast-moving product built by the team behind Saleshandy.
That makes quality trust-critical. An approval that leaks the wrong asset, a reward that fires twice, a count that drifts in the Proof Library those quietly break customer trust. Your job is to make those failures impossible to ship silently.
What's The Role About
Right now, in most teams, verification sits outside engineering manual QA catches bugs in production. This role moves testing left, into the team, so the trust-critical invariants are checked automatically on every PR. Think of it as building the structural safety net, not a manual-testing seat.
You'll write E2E and API tests in Playwright, model flows with BDD and the Page Object Model, validate data with SQL, and wire it all into CI so a broken invariant fails the build before it ever reaches a user. You'll test the real surfaces campaign flows, the approval queue's state transitions, AI-drafted content, notification delivery, and the Proof Library's data integrity.
Three Things Matter More Than Years Of Experience
- High agency. You don't wait for a bug report you go hunt the failure mode before it ships, own the flaky test nobody wants, and stay on the hook for the quality of what you cover.
- Systems thinking. You think in invariants and failure modes what must always be true (a count that can't drift, a state that can't desync, a reward that fires exactly once), what breaks at 100k rows, where partial failures hide. You decide what's worth automating and gating versus leaving manual.
- AI fluency. You use AI to generate tests, fixtures, and edge-case lists fast and you review what it writes, because a shallow test looks fine and catches nothing. AI is a multiplier for your judgment, not a replacement for it.
- Purpose: You're the quality backbone of an early product the reason customers trust what ships. Your safety net protects real users from day one.
- Growth: Learn test architecture, CI gating, and data-at-scale fixtures by building them for real, next to senior engineers who review to teach.
- Leverage: The harness you build gets used by the whole team. We're serious about AI-first engineering, and you'll build a testing workflow that makes you and everyone around you measurably faster every quarter.
- Ship your first automated tests (within 30–45 days)
Outcome: green, non-flaky tests running on every PR.
- Own a test surface end-to-end (within 60 days)
Outcome: a feature surface with automated coverage you maintain.
- Harden the gates (within 60–90 days)
Outcome: measurably fewer escaped bugs and faster feedback on your surface.
- Own quality for a component (within 90–120 days)
Outcome: clear ownership of quality for one surface, and a reusable harness others build on.
Culture Fit Are You One of Us
- You're hungry to learn and grow fast.
- You care about real coverage over vanity metrics, a test that always passes but catches nothing is a red flag to you.
- You take feedback well and give it constructively.
- You move fast, ask sharp questions, and raise a quality risk early instead of letting it slip to prod.
- You use AI to generate and speed up tests, but you verify them you know a shallow test looks fine and catches nothing.