𝗨𝘀𝗶𝗻𝗴 𝗔𝗜 𝘁𝗼 𝗙𝗶𝗻𝗱 𝗘𝗱𝗴𝗲 𝗖𝗮𝘀𝗲𝘀 𝗕𝗲𝗳𝗼𝗿𝗲 𝗖𝗼𝗱𝗶𝗻𝗴

𝗨𝘀𝗶𝗻𝗴 𝗔𝗜 𝘁𝗼 𝗙𝗶𝗻𝗱 𝗘𝗱𝗴𝗲 𝗖𝗮𝘀𝗲𝘀 𝗕𝗲𝗳𝗼𝗿𝗲 𝗖𝗼𝗱𝗶𝗻𝗴

New features feel simple in your head. You picture the happy path. The user does the right thing. The system responds. The page updates.

Real workflows are messy. Users click buttons twice. Two people claim one shift. A user responds late.

I use AI to find these gaps before I write code.

I do not let AI make product decisions. I use it as a reviewer. I ask it to walk through the flow step by step.

I ask these questions:

• What happens before the click?
• Who needs to know the result?
• What happens if the action fails?
• What happens if two people act at once?

These questions force the feature to be specific. Specific features are easier to build.

AI suggests many rules. Some are too complex. I sort them into three groups:

• Must handle now.
• Use a simple default.
• Ignore until users ask for it.

This process improves the build.

• Database models become clear.
• Status names become specific.
• UI actions become narrower.
• Error messages become helpful.

AI does not replace product thinking. It makes product thinking harder to avoid.

It helps you move from a clean idea to a realistic workflow. Your problem decides the product. Not the prompt.

Source: https://dev.to/miran969/how-i-use-ai-to-find-edge-cases-before-building-a-feature-39ik