@LinearUncle: Last week, while communicating with another company, I gave their testing team a small suggestion: Testing must include…
Summary
The author suggests that bug testing should include screen recordings, with AI using ffmpeg to extract clips and analyze frames, significantly improving bug fix success rates.
Similar Articles
@RayFernando1337: The bugs that cause churn almost never show up in a diff, and you only really catch them when you stop reviewing code a…
A developer shares a workflow using Cursor's Opus 4.8 Max Thinking model with subagent harness, and introduces a GitHub repository with installable skill files for AI coding agents, including a 'running-bug-review-board' skill that performs live QA testing.
@gabriel1: every PR will obviously come with 100% coverage of AI app testing, that tries every button in the interface to make sur…
A tweet argues that AI app testing should be a first-class feature in coding apps, noting that many obvious problems could be caught if AI tried the app itself.
@svpino: Instead of reading your AI-generated code by hand, spend time on a plan to verify that your system is working as intend…
The article promotes using Replay QA for automated testing of web applications, especially for AI-generated code, highlighting its ease of use and features like continuous QA and root-cause analysis.
@copyconstruct: end-to-end testing > unit tests, in the vibecoding era. A massive, almost entirely agent-coded refactor passed all unit…
A developer argues that end-to-end testing is more important than unit tests in the age of AI coding agents, after an agent-coded refactor passed all unit tests but broke a critical feature that was only caught by manual end-to-end checks.
The worst bugs in my AI app aren't crashes. They're features that are built, wired, tested, green in CI, and never actually do anything in prod. I've found 20+ of them. How do you catch these?
The author describes frustrating bugs in AI applications that are built and tested but fail to execute effectively in production, such as features targeting wrong users or not firing due to environmental conditions, and asks for methods to catch these issues.