Quality Engineer (all genders) – AI-Augmented, End-to-End Quality
eschbach
·Today
- Location
- Bad Säckingen, DE
- Type
- Full-time
- Department
- Engineering
- Source
- Personio
Description
We are looking for
A Quality Engineer who builds testing AI-native from the start, not as an add-on. You own the quality of our Seqonis applications end to end, from the first requirement to the running product at the customer, and your main tool is Claude. Claude Code and AI agents do the heavy lifting: they design, generate, run, maintain and analyse tests. You set the strategy, direct the agents and own the judgement. As the last check before a release reaches the customer, you make sure our products work the way operators, shift leads, planners and plant managers actually use them.
Your tasks
- Run testing with Claude as your main tool. You use Claude Code as your daily working environment to plan tests, write and refactor test code, generate test data and analyse results. You hand work to agents wherever they do it faster, and you keep the parts that need human judgement.
- Build an agentic test setup. You set up and grow our AI testing toolkit: reusable Claude skills for recurring test jobs, subagents for parallel test design and review, MCP connections (e.g. Playwright MCP, Jira) so agents can explore the application and file findings, and hooks that keep quality checks automatic.
- Turn requirements into tests with AI.You let Claude derive testable scenarios from requirements and acceptance criteria, including realistic workflows across modules, edge cases and failure paths. Then you sharpen and approve them.
- Let agents explore the product. AI agents drive the browser through MCP, walk through complete user journeys, find unexpected behaviour and draft reproducible bug reports for you to check.
- Keep an AI-maintained end-to-end test suite. You build and maintain automated UI and API tests (e.g. Playwright with TypeScript) for critical workflows across modules, APIs and integrations (e.g. OT/historian data, SAP interfaces). Coding agents keep the suite up to date when the product changes, and the suite runs in the CI/CD pipeline with fast, reliable feedback.
- Analyse failures with AI. Agents sort test failures, read logs, tell flaky tests apart from real defects and suggest root causes, so you get from a red pipeline to a clear answer in minutes.
- Own the judgement. You review everything agents produce for correctness, coverage gaps, false positives and hallucinations. AI does the work faster; you are accountable for its quality.
- Test AI-powered product features such as assistants and smart recommendations. Their outputs vary from run to run, so you evaluate them with defined quality criteria, scoring rubrics, LLM-as-judge setups, human review and clear pass/fail thresholds.
- Define a risk-based test strategy across all test levels (unit, API, integration, end-to-end), with end-to-end coverage of business-critical user journeys as the guiding goal, and decide where AI, automation or manual testing fits best.
- Test manually and exploratively where human judgement adds value: usability, new features and complex shop-floor scenarios.
- Drive measurable quality. You keep the defect escape rate low and test runs stable and free of flaky tests, and you track how much AI speeds up testing and where it falls short.
- Spread AI-native testing in the team. You share prompts, skills and workflows with developers, the Squad Lead, the Product Manager and Customer Solutions, so quality and testability are part of the work from the start.
- Use AI responsibly. You handle test data and AI tools in line with our data protection and information security rules, using synthetic or anonymised data and approved tools only.
Your profile
Must have
- Several years of experience testing web applications, both manually and automated
- Daily, hands-on work with Claude (ideally Claude Code) or comparable AI coding agents to write, maintain and analyse tests. You can show concrete examples of what you built with them.
- Experience with agentic and MCP-based test automation (e.g. Playwright MCP), and with setting up agent workflows such as skills, subagents or custom instructions
- Strong experience in end-to-end test automation for UI and API (Playwright preferred; Cypress or Selenium welcome) with TypeScript/JavaScript or Python
- Experience integrating automated tests into CI/CD pipelines (e.g. Azure DevOps, GitHub Actions)
- Strong skills in prompting and context design: you know how to give an agent the right context, check its output and correct it
- Strong skills in exploratory testing and risk-based test design
- Experience with SQL databases and with tools such as Jira for bug tracking and test management
- Very good English; German is a plus
- Experience testing AI/LLM-based features (non-determinism, evaluation frameworks, LLM-as-judge, prompt and output validation)
- Experience building your own MCP servers, Claude skills or agent pipelines
- Knowledge of performance and security testing
- Background in industrial software, manufacturing or regulated environments (pharma, chemical)
Our offer
- AI as standard – modern AI tools built into your daily sales work.
- Great team spirit – Work with colleagues who make collaboration fun and always have each other’s backs.
- Personal development – We support your strengths and give you the space to grow both professionally and personally.
- Feel-good workplace – Enjoy a modern, ergonomic office in the heart of Bad Säckingen, complete with a rooftop terrace, free parking, and plenty of cozy spots to work or relax.
- Team culture that connects – BBQs, Christmas parties & more – we love spending time together beyond work.
- Flexible working – Thanks to our home-office policy, you stay connected and productive even when working remotely.
- Snacks & soul food – Fresh coffee, fruit, and snacks are always available – plus, you can enjoy free, delicious lunch options from local spots in town.
Ready for the challenge?
Interested in joining our team?
Apply directly by clicking the "Apply for this job" button!
Contact person
Do you have questions? Contact Kira Hinkelmann (E-Mail: [email protected])!