Kai Long

English isn t my first language, and I ve noticed something odd while vibe coding: I get better results when I prompt in English, but I think more clearly when I write in my own language.

A few things I ve run into:

• When I describe a UI or a bug in my native language, the AI sometimes picks up the intent fine, but gets technical terms wrong or mixes languages in the output.

• Prompting in English feels safer for code, but I lose nuance, especially when I m explaining what a feature should feel like rather than what it should do.

• Translating my own thoughts before every prompt adds friction, and I catch myself writing shorter, vaguer prompts just to avoid it.

So I m curious how others handle this:

Getting a working app out of a few prompts is the easy part now. What I keep running into is the day after: the app works, but I don't really know what's inside it.

A few things that bit me after the "wow, it runs" moment:

• Realizing I couldn't explain how auth or data storage was actually set up

• Fixing one bug and breaking something unrelated because I never saw the structure

• Coming back a week later and not recognizing my own project

So I've started doing a short "day 2 pass" before I build anything new on top: skim the structure, check where secrets and user data live, and write a few lines about what the app does and how.

AI can get me to a working app in a day, but "it works on my machine" and "it's safe for real users" are very different things.

The part that worries me is the boring stuff that's easy to miss when the code looks clean and everything runs: API keys sitting in the frontend, auth checks that exist on the page but not on the endpoint, database rules left wide open from the prototype stage, no rate limits on anything. The AI rarely warns you about these unless you ask, and by the time you think to ask, you've already moved on to the next feature.

Right now my pre-launch process is mostly manual and a bit random: I skim through the auth flow, search the repo for hardcoded secrets, and try to break things as a logged-out user. It catches some things, but I never feel fully confident.

I'd love to hear how others handle it: