Building a Small Niche Mobile Game on Your Own With AI Coding Help, Monetized Through Optional In-app Purchases
Originally published Sep 27, 2026 — Starter Story
Starter Story · This app replaced my 9-5 ($155K/year) ↗
Build a small free-to-play mobile game by yourself for a specific niche of players you already understand, using AI coding tools (Cursor, ChatGPT) to write most of the code. Make money through optional in-app purchases that unlock premium features or speed up progress. The featured developer, a former Meta engineering manager who hadn't coded in four years, built a text-based idle RPG for fans of incremental games like RuneScape. He bought ready-made art from asset stores (Unity Asset Store, GameDev Market), hired freelancers on Fiverr/Upwork for icons, screenshots and videos, and ran the backend on Firebase.
The developer reports about $13,000/month on average (over $150,000 in the past year) from only 3,500-4,000 monthly active players, with roughly 50,000 total downloads, a 6% paying conversion rate and about $3 in revenue per active user. His argument is that you don't need a hit game: a small audience that genuinely loves a niche genre can support a full-time income. He spent nothing on ads. His revenue jumped from about $1,000 to $10,000 a month after a single Reddit post, and he found that posts drawing the most criticism brought in the most new players. He also found in-app prices between $4.99 and $9.99 convert best.
Someone who plays and deeply understands a particular game genre, is comfortable learning to build software with AI help, and can accept a slow, uncertain start. It suits people who can keep a day job while testing, as the developer did before quitting.
The figures are self-reported in an interview on a channel that promotes indie-builder success stories, and they weren't shown in a way that can be independently checked, though the game itself is real and publicly available. This is one success case; most indie mobile games earn little or nothing, and his result depended heavily on one Reddit post taking off, which he says wasn't consistent. His 12 years as a software engineer at Amazon and Meta almost certainly made the AI-assisted build far easier than it would be for a true beginner. The strategic advice (build for a niche you know, launch early, focus on keeping players before monetizing) is sound and widely shared among indie developers.