Quality Assurance Engineer*

Oct 24, 2026 Oct 25, 2026

BAWAG Group Quality Assurance Engineer*

About the Role

We're looking for a QA Engineer who doesn't just find bugs — they prevent them.
You'll be embedded in a cross-functional product team, working closely with developers and a Product Owner to build quality in from day one. You write code, you break things on purpose, and you use AI to work faster and smarter.
One thing that makes this role different: we already run an AI agent pipeline on the product side. Ideas get shaped, analysed, broken down and turned into Jira tickets by a chain of AI agents with a human reviewing every gate. It works, and it's changed how fast we move.
What we don't have yet is the equivalent on the quality side. That's yours to build.

**The QA craft**

Design and maintain automated test suites (E2E, integration, regression) with Playwright as the primary tool
Write and execute test strategies for Angular frontends and Java backends
Build and own CI pipeline quality gates — no broken build makes it to staging on your watch
Perform API testing (REST, Postman/Newman, contract testing)
Contribute to test data management and environment stability
Conduct exploratory testing focused on edge cases and real user journeys
Triage defects, reproduce them reliably, and communicate impact clearly
Champion quality practices across the team — you're the one who raises the bar

**The AI side**

Extend our existing AI agent pipeline into QA — the same way we generate requirements today, we want to generate test plans, test cases and Gherkin specs that are actually good enough to use
Plug into the pipeline upstream: acceptance criteria written by an agent should come out testable, and you're the one who makes sure they do
Use AI to expand coverage, hunt edge cases a human wouldn't think of, and keep test suites from rotting
Build the prompts, context files and guardrails that make agent output reliable rather than plausible-looking
Decide what AI should *not* touch — some things need a human, and knowing where that line sits is part of the job
Review everything. AI output that ships unreviewed is a defect waiting to happen, and you own that standard for the team

**5+ years in QA** with a strong automation focus
**Playwright** — you've built and maintained E2E suites from scratch; you understand selectors, fixtures, page object models and parallel execution
**SQL** — you write queries confidently, understand joins and indexes, and validate data at the source without asking a developer
**AI-assisted testing** — you actively use AI tools (Copilot, Claude, ChatGPT or similar) in your day-to-day. You know what a good prompt looks like, you've been burned by a bad one, and you can tell the difference between output that's correct and output that just reads well
Solid understanding of Angular applications and how to test them (component, E2E, snapshot)
Comfortable reading and writing Java or TypeScript — you're not just clicking around
Experience testing REST APIs — HTTP, auth flows, JSON schemas
Familiarity with CI/CD pipelines (Jenkins preferred — that's what we run)
Knowledge of test pyramid principles — you know when *not* to write an E2E test

## Strong Plus

**Power BI** — you build dashboards and reports the team actually uses (quality metrics, coverage, defect trends), not ones that get opened once
**Jira + Xray** or similar test management tooling
Experience in **banking, fintech or another regulated industry** — you understand why the audit trail matters
You've built something with AI agents before — custom instructions, agent workflows, MCP, RAG, whatever it was. Tell us what worked and what didn't

## Nice to Have

## What We Value

You ask **why** before you test what
You treat flaky tests as bugs, not noise
You're genuinely curious about AI, but you're not a believer — you've seen it get things confidently wrong and you check the output
You give honest, constructive feedback in code reviews
You document just enough — clear, not exhaustive
You care about the end customer, not just the ticket

Skills