An automation tester writes and maintains tests that a computer runs instead of a person. The goal is not to replace manual testing, but to take over checks that repeat in every release.
What an automation tester does
Chooses which test cases are worth automating: stable, repeated in every release and expensive to run by hand.
Builds automated tests and keeps them working when the application changes.
Prepares test data and environments in which tests can run every time.
Reads the results of runs, separates real defects from broken tests and reports them to developers.
Connects tests to the delivery pipeline, so a release does not go out with known regressions.
Skills the role needs
Knowledge of testing itself: test design, boundary values, equivalence classes, risk.
A tool for browser automation such as Playwright or Selenium, and basic programming in JavaScript, TypeScript, Java or Python.
Selectors (CSS, XPath, roles) and an understanding of how web pages load.
Version control with Git and basics of CI tools.
Patience and accuracy: an unstable test costs more than no test.
Manual and automated testing together
Automation is good at repeating the same checks quickly. People are better at exploring, judging usability and noticing what nobody planned to check. Most teams run both: automated regression tests for the stable core and manual tests for new features and edge cases.
Starting without a dedicated automation engineer
A team without programmers can still automate the most repetitive checks. In Testosaur an automated test case is built from ready steps, runs with Playwright and sits in the same test plan as manual tests, so the team gets results and screenshots without maintaining a separate framework. Read also about the manual tester role.