How I Built My First Profitable App

After being fired from his corporate job, a software engineer decided to build an AI-powered calorie tracker for iOS from scratch without prior knowledge of Swift. By focusing on a "pay-as-you-go" AI model and navigating Apple's rigorous App Store hurdles, he successfully launched the app and achieved organic revenue in its first week.

---

1. From Corporate Layoff to Indie Developer

Life took a sudden turn for Andy when he was let go from his corporate software engineering position. Instead of looking for another 9-to-5, he set an ambitious goal: building his first-ever iOS app and documenting the entire journey. Despite his background in web development and APIs, he had zero experience with Swift, Apple's native programming language.

Inspired by the thriving indie developer scene on X (formerly Twitter), he aimed not just to build an app, but to create a sustainable business.

"I saw a lot of solo iOS app founders hit astronomical numbers with their apps, much more than my job salary could ever get to."

The idea for his project came from his personal passion for fitness. He decided to build an AI-powered calorie tracker where users could simply snap a photo of their meal to get instant nutritional data. To make it stand out from the competition, he added a "twist": tracking how different foods affect the user's energy levels and feelings throughout the day. 🥗

---

2. Choosing the Right AI and Tech Stack

Building an app that recognizes food requires powerful AI. Andy explored several options, including training a custom model (which was too time-consuming) or using high-end APIs (which were too expensive, costing upwards of $100/month).

He eventually settled on a "pay-as-you-go" microservice. This model charges only for the images analyzed, making it much more affordable for a solo developer starting out. Using the Cursor code editor (an AI-powered IDE), he began setting up the backend and the iOS simulator.

"The best solution I landed on was the pay-as-you-go microservice. This one will charge you for the pics it analyzes and seems to have a pretty low price as well."

He faced immediate hurdles with the backend, struggling to connect the AI model to the database and get accurate responses. At one point, he nearly gave up, but a breakthrough occurred when the AI successfully identified a yogurt cup, proving the concept was viable. 📈

---

3. Designing for User Experience (UX)

With the core functionality working, Andy shifted his focus to the "identity" of the app. He used platforms like Dribbble and Mobbin for design inspiration and Figma for creating iPhone mockups. He emphasized that for an app to succeed, the onboarding process and the paywall are the most critical sections.

• Onboarding: He spent a significant amount of time ensuring the first-time user experience was smooth, as this is where most users are lost.
• Authentication: He implemented Google and Email logins. Interestingly, he initially used "magic links" for email login, which caused a major headache later during the Apple review process.
• Analytics: He integrated PostHog to track exactly where users might be dropping off during the onboarding flow.

"I watched podcasts from Superwall and Starter Story and they were all saying how important the onboarding and paywall is. So, I took my time with it."

---

4. Navigating the "Elite" Apple Developer Club

To actually publish an app, you must join the Apple Developer Program for $100 a year. Andy described this process as "riddled with bugs," including an interface that made him enter his payment details twice.

Once inside, he faced the confusing world of App Store Connect. Configuring subscription packages turned into a nightmare of "missing metadata" errors. The solution wasn't found in documentation or through AI, but in a "dusty subreddit." It turns out Apple requires specific "groups" for subscriptions that aren't clearly explained in the UI. 🍎

"Least they could have done was a clear error message about this somewhere, but no. Took me half a day to figure this out."

---

5. Rejection and Eventual Success

After a month of hard work, Andy submitted his app, BiteTracker AI, for review. He was promptly rejected. Apple's reviewers found a critical bug: the paywall was appearing even for users who had already paid, effectively locking them out of the app.

"It would literally not let the paid user use the app. Thank god Apple tested that flow."

He fixed the bug, resubmitted, and finally got the green light. In its first week, the app generated organic revenue—money earned without any paid advertising—partly thanks to Apple's initial "boost" for new apps. 💰

---

Conclusion

Building a profitable app is less about master-level coding and more about persistence and problem-solving. By leveraging AI tools like Cursor and focusing on a proven niche with a unique twist, Andy transitioned from a laid-off engineer to a revenue-generating indie developer. His next step involves marketing through social media to scale the app to five-figure monthly recurring revenue.

"This was a bit of a longer one. So, thank you guys for sticking with me up to the end. And until next time, happy coding."