Senior Software Quality Assurance Engineer

Location: Australia (Remote, with occasional client-site or hybrid presence) Experience: 8–12 years in software testing, including 5+ years building automation from scratch Engagement: Freelance / Contract / Consultant — full-time or part-time, flexible Team: Small, senior, product-and-services hybrid. You will be one of a handful of people, not one of a hundred. About the roleWe build and migrate enterprise web platforms — Sitecore XP and XM Cloud, Drupal 10/11 — for clients who notice when something breaks. Right now, quality is owned by the developers who wrote the code. That has a ceiling, and we have reached it. We are hiring the person who owns quality end-to-end: the test strategy, the automation framework, the exploratory testing that finds what automation never will, and the conversation with the client about whether we are ready to go live. "Full stack" here means the whole depth of the system — browser UI, REST and GraphQL APIs, database and content integrity, mobile, performance, accessibility and security. What you'll actually doOwn quality, and own the client conversation● Define test strategy per engagement — scope, risk, environments, entry and exit criteria, what we automate and what we deliberately do not ● Present test plans, coverage, and release readiness directly to client stakeholders, often without a PM in the room ● Give an honest go/no-go recommendation and defend it when there is pressure to ship ● Estimate QA effort for proposals and hold the line when scope moves ● Write reports non-technical stakeholders can act on — defect trends, risk areas, coverage gaps Manual and exploratory testing● Risk-based exploratory testing on new features, with session notes that stand up to review ● CMS authoring and editorial workflow testing — content modelling, workflow states, personalisation rules, multilingual behaviour, publishing and preview. Editors are users too, and their experience is routinely untested. ● Migration validation — one of our highest-risk activities. Content counts and integrity, media and asset fidelity, URL and redirect mapping, metadata and SEO preservation, permissions, taxonomy accuracy. Build repeatable comparison tooling rather than spot-checking spreadsheets. ● UAT support: prepare scenarios, run sessions with client teams, triage what comes back ● Sharp defect reports — reproducible, with evidence, severity, and business impact Build and own the automation framework● Design and build automation from scratch: architecture, patterns, data management, environment strategy, reporting ● Rapise 9.0 — scripted and scriptless tests, SmartActions and AI-based visual recovery for resilient locators, image-augmented validation during playback ● Selenium 4.x, Playwright, Cypress, Appium — you should be fluent in code-based automation, not only low-code tooling ● Build for maintainability. A suite nobody trusts is worse than no suite. ● Set and defend the automation/manual boundary rather than automating everything on principle Use AI tooling with judgement● Rapise 9.0 self-healing — SmartActions combining object description with natural-language intent, AI visual recognition to recover from broken locators, multi-level recovery ● Inflectra.ai — text and image queries during execution for visual validation and in-flow decisioning ● BrowserStack AI agents — Low-Code Authoring (natural language and requirements into tests), Self-Healing Agent, Test Case Generator, Test Failure Analysis, Test Deduplication, Visual Review in Percy, and A11y Detection and Remediation ● LLM-assisted test case design, test data generation, and defect triage ● Critically: know the failure modes. AI-generated test cases drift toward the happy path. Self-healing can mask a genuine regression by quietly adapting to a bug. Visual diff filtering can suppress a real change. We want the productivity, and we want someone who audits it. Cross-browser, cross-device and mobile● BrowserStack Live, Automate, App Automate, and Percy across real browser and device matrices ● Define the browser/device matrix from the client's actual analytics, not a generic list ● Responsive and native mobile testing; iOS and Android real-device coverage ● Visual regression testing that flags real breakage without drowning the team in noise APIs, data and performance● REST and GraphQL testing — contract, schema, auth, error handling, edge cases (Postman/Newman, REST Assured, or equivalent) ● Integration testing across CRM, search, commerce, SSO, and third-party services ● Database validation and data integrity checks, particularly post-migration ● Performance and load testing (JMeter, k6) and front-end performance against Core Web Vitals ● Security-adjacent testing: authentication, authorisation, role and permission matrices, input handling Accessibility● WCAG 2.2 AA testing as a delivery standard — a contractual requirement for many of our government, education and nonprofit clients ● Automated scanning (axe, Lighthouse, BrowserStack Accessibility) plus manual keyboard navigation and screen reader testing (NVDA, VoiceOver). Automated tools catch roughly a third of issues; the rest need a human. ● Produce audit reports and remediation guidance developers can act on Build the QA practiceThis role is the QA practice, and you are building it. ● Establish standards, templates, and tooling decisions across all engagements ● Set up test management properly — SpiraTest/SpiraPlan, BrowserStack Test Management, or Jira with Xray/Zephyr — and make traceability real rather than decorative ● Integrate testing into CI/CD as enforced quality gates (GitHub Actions, GitLab CI, Azure DevOps, Jenkins) ● Build reusable accelerators — migration validation harnesses, CMS regression packs, accessibility baselines — that shorten every subsequent project ● Turn QA into a service line. Standalone test automation, accessibility audits, migration assurance, and performance audits are sellable engagements. Help us scope, price, and pitch them. ● Mentor developers into better testing habits and bring on junior QA as we grow RequirementsCore testing (mandatory)● 8–12 years in software testing with genuine depth in both manual and automated approaches ● 5+ years building and owning automation frameworks — architecture decisions, not just writing tests in someone else's structure ● Strong test design technique: equivalence partitioning, boundary analysis, decision tables, risk-based prioritisation ● Exploratory and session-based testing you can evidence ● Defect lifecycle management and clear, credible reporting Tooling● Rapise — hands-on experience strongly preferred; Rapise 9.0 AI features a clear differentiator. Deep experience with a comparable tool plus willingness to convert quickly is acceptable. ● BrowserStack — Live, Automate, App Automate, Percy; AI agent features preferred ● Selenium 4.x, Playwright, Cypress, Appium — fluency in at least two ● API testing: Postman/Newman, REST Assured, or equivalent; GraphQL testing ● Performance: JMeter, k6, or equivalent ● Accessibility: axe, Lighthouse, WAVE, plus real screen reader use ● Test management: SpiraTest/SpiraPlan, Xray, Zephyr, TestRail, or BrowserStack Test Management Technical● Comfortable in at least one of JavaScript/TypeScript, Python, Java, or C# — you will write real code ● SQL for data validation ● Git and branching workflows ● CI/CD pipeline configuration — you own your own pipeline integration ● Reading application logs and browser dev tools well enough to diagnose rather than just report ● Docker exposure (good to have) Domain● Experience testing enterprise CMS or DXP, Insurance platforms — Sitecore or Drupal a strong plus; any enterprise CMS is relevant ● Content migration validation experience — highly valued given our project mix ● WCAG 2.2 AA in practice ● ISTQB Advanced Level, or equivalent demonstrable depth — preferred, not mandatory Who this suits● You are as proud of a well-designed exploratory session as of a green pipeline ● You have opinions about what not to automate, and can defend them ● You can tell a client "this isn't ready" and keep the relationship intact ● You treat AI tooling as a capable assistant to be supervised, not an oracle ● You are comfortable with ambiguity, thin process, and shifting priorities, because that is the reality of a small team ● You context-switch across two or three clients in a week without dropping threads ● You write clearly. Half this job is communication. Who this doesn't suit● Anyone who needs a fully-specced test case handed to them before starting ● "Automation engineers" who consider manual testing beneath them — half this role is manual, by design ● Manual testers who have avoided learning to code ● Anyone whose automation experience is entirely record-and-playback with no framework design ● Anyone who would ship an AI-generated test suite without reading it ● People who report defects but won't help diagnose them