Three weeks ago I pasted a draft of a pretty sensitive professional email into Grammarly. You know the kind — delicate wording, confidential situation. Hit paste, Grammarly lit up with suggestions. I accepted a few.
Then I stopped and thought: where did that text just go?
I went and read Grammarly's privacy policy. It was... fine, actually. They say they don't sell your data. But it still goes to their servers. Every sentence, every draft, every awkward thing you've ever typed into an email — it all leaves your device.
That bothered me more than I expected.
I'm a backend dev by trade. I've been writing APIs and distributed systems for 8 years. I understand what "processing on our servers" means at a technical level. And I also know that "we don't sell your data" is not the same as "your data never leaves your phone."
Around the same time, I started reading about Apple's on-device AI — the Foundation Models framework that ships with iOS. It's designed exactly for this: inference that runs locally, no network calls, no data sent anywhere.
A grammar checker that runs entirely on your phone. No server. No account. No copy of your text on someone else's infrastructure.
That felt worth building.
What Proofed is
A grammar checker for iPhone that works offline. That's it.
- Checks grammar and spelling
- Suggests rewrites (Pro)
- Adjusts tone (Pro)
- Works with no internet connection
- Requires no account
- Stores nothing
The AI is Apple's. I'm just wiring it up in a way that makes it useful for people who write on their phones.
What I'm trying to figure out
I've never shipped an app on the App Store. I've built plenty of internal tools and APIs, but consumer apps are different. Distribution is different. Monetization is different.
So part of this is also a learning exercise for me. I'm going to document what works and what doesn't — pricing experiments, ASO mistakes, which marketing channels actually moved the needle, all of it.
If you're building something similar, hopefully some of this is useful. If not, at least it'll be an honest account of one guy's attempt to ship an indie app.
The plan
Build the MVP in a few weeks. TestFlight in March. App Store submission once it's stable.
I'm not trying to kill Grammarly. I'm building for the specific person who wants their writing checked but doesn't want it leaving their device. That person exists. I'm that person.
Follow along on X @lucas_merr if you're into this kind of thing.