I built a tool to translate JSON i18n files in seconds — after ChatGPT kept breaking my production files

A few weeks ago I ran into a problem that was small, but incredibly frustrating.

I was using AI to translate JSON i18n files. It worked, until it didn’t.

Keys would get changed.
Placeholders like {name} or %d would break.
Sometimes the JSON wouldn’t even parse.

Every time, it meant manually fixing the output before I could use it.

That defeats the whole point.

The real issue

AI is great at generating text.
It’s not designed to respect strict structure.

And i18n files are all about structure:

keys must stay identical
placeholders must remain untouched
formatting must be valid

If any of those break, your app breaks.

So I built a simple tool

It does one thing:

Translate JSON i18n files without breaking anything

You upload a file, choose a language, and get back:

same keys
placeholders preserved
valid JSON
ready-to-use output

No cleanup needed.

What surprised me

Building it was straightforward.

The interesting part was testing it against real files.

deeply nested objects
arrays
mixed placeholders ({name}, %d, {count})
UI strings, emails, notifications

That’s where most tools fail quietly.

Where I need feedback

Right now I’m trying to understand:

Is this something you’d actually use in your workflow?
Where does this fit vs existing tools/processes?
What would make this a “must-use” instead of a “nice-to-have”?
The tool

If you want to try it:
https://localizejson.lovable.app/

Still early, but I’d rather get real feedback now than overbuild.