July 24, 2026
Prompt, Pray, Repeat
I Tried Building a Real App with AI. It Took a Year
AI was supposed to make app-building easy — commenters say reality had other plans
TLDR: A developer spent a year building a habit app with AI after cheaper existing apps didn’t meet his needs. Commenters turned it into a bigger debate over whether AI is truly a shortcut or just another messy coworker that still needs a human to finish the job.
A developer set out to build a simple habit-tracking app for himself and his wife after getting fed up with pricey existing options and apps that just didn’t fit. On paper, this should have been the ultimate “AI does it while you sip cocktails” moment. In reality? It turned into a year-long slog, and the comments section absolutely pounced on that gap between the fantasy and the messier truth.
The loudest reaction was basically: making real software is still hard, sorry! One commenter bluntly said good software still takes time, skill, and knowing what you’re doing, with artificial intelligence acting more like a power tool than a wizard. Another came in with a spicy reality check, asking how much of the app was actually written by hand after the first few hours, because in their experience, a weekend of prompting can already get something usable. That sparked the classic internet split: is AI a huge shortcut, or just a machine that creates extra cleanup?
And yes, the jokes wrote themselves. One of the funniest lines compared cleaning up AI-made messes to cleaning up human-written slop since 2003 — which is both a roast and a cry for help. Another big mood in the thread was nostalgia-fueled outrage over app pricing, with the original story’s complaint about habit apps costing almost as much as a game landing hard. The overall vibe? AI can help, but if you thought it could replace the boring, fiddly, deeply human grind of making something polished, the community is here to laugh, nod, and say: welcome to the pain.
Key Points
- •The author and his wife could not find an existing habit-tracking app that matched their preferred feature set and pricing.
- •Required features included a long-list interface, local-only storage, custom iOS 18 gen-emoji icons, flexible goals, occurrence counting, notes, and reminders.
- •Streaks met several needs but lacked support for occurrence counting, preferred layout, icon flexibility, and disabling streaks.
- •HabitKit and Grit were also evaluated, but the author found their pricing too high relative to what was needed.
- •The author began building a custom app with AI assistance in March of the previous year, using free credits from Cursor.