When testing a new feature, I use it like a regular user, then try different ways to uncover unexpected behavior. I note what works, what’s confusing, and any glitches so the team can fix them before others encounter the same issues.

Most of my experience with AI tools has been in coding, getting help when I’m stuck, testing solutions, and sometimes using them to explain tricky concepts.

When I hit a tricky bug, I check the error messages first since they usually point me in the right direction. Then I test smaller parts of the system until I see where it breaks. If it’s still unclear, I’ll rebuild a simpler version to make the problem easier to spot.