Skip to content

Apple approved my teleprompter app in 8 hours. An AI submitted my game. 29 cold emails: 0 replies.

Yesterday I said I'm an engineer with 6 products and $0, and that for 6 months I'm going to learn distribution in public. Day 2 was the clearest proof of my own point I could have asked for.

Cracked Marketer, day 2 of 180: Shipping is solved. Getting one reply isn't. 8h 42m Apple approval, 129 commits by AI, 29 cold emails, 0 replies.

Yesterday I said I'm an engineer with 6 products and $0, and that for 6 months I'm going to learn distribution in public.

Day 2 was the clearest proof of my own point I could have asked for.

In one day, Apple approved one of my apps and an AI submitted another one for me. And the one thing that actually makes money, people, went nowhere.

Here's the day, with the real numbers.

Win 1: Apple approved FoldPrompt in 8 hours 42 minutes

FoldPrompt is a teleprompter for foldable iPhones. You read your script on one screen while the camera looks at you from the other.

  • Submitted: 8:02 AM.
  • Approved: 4:44 PM. Same day.
  • First try. No rejection.

Two days earlier another app of mine took about 38 hours. A third one got rejected this week and is back in the queue.

What changed? Nothing clever. Clean build, real screenshots, an honest privacy label ("Data Not Collected"), 35 automated tests passing. No tricks. Boring is fast.

It's live on the App Store now, free in 174 countries: apps.apple.com/us/app/foldprompt-teleprompter/id6832142416

Win 2: An AI built my game and submitted it to Apple (not approved yet)

Quick note: I'm a Devin ambassador, so take this with that in mind. But this happened.

Devin built a naval strategy game for a foldable iPhone, end to end. It made the 3D models, wrote the code, ran 119 tests, recorded the simulator and sent me the videos, found bugs I would never catch, uploaded to TestFlight and pressed Submit for Review.

129 commits. Every one of them by the AI. I never opened Xcode.

The game is still waiting for review. Not approved yet. I'll tell you what Apple says.

The loss: 29 cold emails, 0 replies

Now the part that matters.

Umless is my video tool. It cuts the "um"s and dead air out of your videos. It's the only product I'm focused on for 30 days.

My agents have sent 29 cold emails in 3 days. Here's the scoreboard:

  • Replies: 0
  • Bounced: 3
  • Paying customers: 0
  • MRR: $0
  • Waitlist: 1 person
  • Visits to the site in 24 hours: 144 human hits. 0 came from a link anywhere. All direct.
  • GitHub stars: 1

And my favourite one: my AI account balance hit -$0.16. Sixteen cents. That broke the free trial, the renders and the blog images. The agents kept working, they just couldn't finish anything.

So the machine that ships products is great. The machine that finds people is not working yet.

What I learned on day 2

1. Shipping is solved. Attention isn't.

One app approved in 8 hours, one game built and submitted by an AI, both in one day. That's the new normal. If shipping is that cheap, a shipped product is worth nothing on its own. The scarce thing is a stranger who cares.

2. Cold email from nobody gets nobody.

29 emails from an account with no reputation, about a product nobody's heard of. 0 replies is the honest result. Email works better after people have seen you somewhere first. That's what this account is for.

3. My agents found the angle. I just have to use it.

The research agent read 14 new complaints about a big competitor. 8 of them were about the same thing: paying credits to remove filler words. One person said it cost "800 credits for 1 video". So today the site changed to one line: a Descript alternative with no credits to burn. That's a real pain in real words, not my guess.

4. One person with $5 is the bottleneck.

The -$0.16 thing is funny, but it's the same lesson as day 1. Agents do the work. The human has to unblock it. Top up the account, press send, reply to people. That's my job, not theirs.

What I'm changing tomorrow

  1. Send before build. Every morning, posts and outreach go out before I write a line of code.
  2. Conversations, not emails. The scoreboard is conversations started. Target: 5 a day. Today: 0.
  3. Show up where the buyers already are. Creators complaining about editing time, not other developers.
  4. November 17: Umless goes to Hacker News. That's the first real test.

If you're a founder

Here's the uncomfortable math from my week: I can get an app approved in 8 hours, have an AI build and submit a whole game, and still have 0 customers.

If you have a product nobody knows about yet, reply with the link. I'm picking a few to tear down live on X next Sunday.

Day 2 of 180. Tomorrow I'll tell you if anyone replied.

Every article, with links and sources: thoufeekxlabs.com/articles

Originally on X: read and reply on X

One of these every week. One insight a day.

FollowBook a call