An inside look at software testing at GitLab

What this post added

This post discusses the challenges of software testing, specifically the lack of sufficient tests and the issue of flaky tests. It highlights GitLab's efforts to improve test coverage by emphasizing test requirements in issue and merge request templates. It also touches upon the need for test frameworks to be reliable and mimic manual tester behavior, likening user workflows to retrying dominoes. The post introduces the concept of using machine learning for test automation, envisioning a bot that can randomly interact with UI elements to find errors. Finally, it outlines goals to increase quality and productivity by adding more automated tests for enterprise features while shortening test runtime.