About the role
We're looking for a tester who codes, not a coder who writes scripts. You'll work on-site at the client, testing an LLM-based application. You'll design and build automated tests that reflect how real users behave, not just how the UI is wired. You'll decide what's worth automating, what isn't, and why.
Responsibilities:
- Explore the product to understand users, workflows, risks, and edge cases before writing a single test.
- Design test scenarios from the user's perspective: real journeys, real data, real failure modes.
- Build and maintain a Playwright + TypeScript automation suite that's readable, stable, and easy to extend.
- Write tests that catch meaningful problems, not tests that only confirm the happy path.
- Investigate failures, separate product bugs from test flakiness, and fix the root cause.
- Integrate tests into CI/CD pipelines and keep feedback fast.
- Report bugs with clear impact, steps, and evidence so developers act on them.
- Work closely with developers and product owners to shape testability early.
Requirements:
- Strong hands-on experience with Playwright and TypeScript.
- Solid testing fundamentals: test design techniques, risk-based thinking, exploratory testing.
- Ability to look at a feature and ask "what could go wrong for the user? " before "how do I automate this? "
- Good grasp of page object or similar design patterns, fixtures, and test data management.
- Experience with API testing (Playwright API, Postman, or similar).
- Working knowledge of Git and CI tools (GitHub Actions, Jenkins, GitLab CI, or similar).
- Clear written and verbal communication.
Good to have:
- Exposure to testing systems that use LLMs: non-deterministic outputs, prompt variations, and response quality.
- Familiarity with performance testing basics.
- Experience working at a client site with enterprise stakeholders.