Bitrig Lets You Build iPhone Apps From Your iPhone, and the Founders Literally Created SwiftUI

The Macro: Building iPhone Apps Still Requires a $2,000 Laptop and a Masochistic Tolerance for Xcode

There is a strange irony at the heart of iOS development. The iPhone is the most popular computing device in human history. Over a billion people carry one. But if you want to build an app for it, you need a Mac, you need Xcode (which is notorious for its bloat, slow build times, and cryptic error messages), and you need to learn Swift, a language that is elegant but non-trivial. The barrier to entry for creating something on the platform that defines modern computing is absurdly high.

The “vibe coding” movement changed this equation for web apps. Lovable, Bolt, and Replit let you describe a web app in plain English and get a working product in minutes. Cursor and Windsurf turned coding into a conversation. But all of these tools produce web applications. React, Next.js, HTML, CSS. If you want a native iOS app, one that feels right, that integrates with the system, that uses the camera and sensors and notifications the way users expect, you are back to Xcode and Swift and the full Apple developer toolchain.

This matters because native apps are meaningfully better than web apps on mobile. The animations are smoother. The gestures feel natural. The integration with the operating system is deeper. Users can tell the difference, even if they cannot articulate why. The companies that ship native apps get better reviews, better retention, and better performance. But the cost of building native is so much higher than building for the web that most startups and indie developers default to React Native or Flutter, which are compromises dressed up as solutions.

The market for AI-assisted code generation is exploding. Lovable reportedly hit $30 million in ARR. Cursor is growing at a pace that makes enterprise SaaS companies jealous. But native iOS development has been left out of this revolution because the toolchain is uniquely complex and proprietary. Nobody has built the “Lovable for iPhone apps” because building a Swift code generator that produces apps good enough for the App Store requires deep expertise in both AI and the Apple platform.

The Micro: The People Who Built SwiftUI Left Apple to Let You Skip Xcode

This is one of those founding teams where the background is not just relevant, it is the entire story. Kyle Macomber (CEO) spent over ten years at Apple as a co-creator and lead developer of SwiftUI. He also managed the teams behind Swift Charts, WidgetKit, and the Swift Standard Library. Jacob Xiao co-created and led SwiftUI development over 15 years at Apple, with additional work on Xcode and UIKit. Tim Donnelly worked on SwiftUI at Apple, then Square, co-founded Storehouse (which won an Apple Design Award and was acquired by Square), and rewrote the fluid animation system that runs on 1.5 billion iOS devices. He has a BFA in Design from the University of Washington.

These are not people who learned Swift from a tutorial. They built the framework that every iOS developer uses. When they say they can generate production-quality SwiftUI code from natural language descriptions, the claim carries weight that it would not carry from anyone else on earth.

Bitrig (Y Combinator Summer 2025) lets you describe an app in plain English, and it generates real Swift and SwiftUI code. Not pseudo-code. Not a prototype. Actual, readable, editable code that you can submit to the App Store. The app includes an integrated iPhone simulator that runs alongside the code editor, so you see your changes in real time. You iterate by having a conversation: “make the header blue,” “add a settings page,” “integrate the camera.”

The key technical achievement is a custom Swift interpreter they built from scratch that supports native SDK access. This is not a wrapper around a web view pretending to be native. The code it produces uses the actual iOS APIs, the actual SwiftUI components, the actual system integrations. The app runs natively, which means it performs like a native app, because it is one.

The product requires macOS Sequoia 15+ for the desktop version, and it is available on the App Store. The testimonials are telling: a non-technical founder who successfully launched an app, a blind user who called it “extremely good” for accessibility, a developer returning after a 40-year hiatus. These are the people that native development has excluded, and they are building real apps.

The competitive landscape is thin for native iOS. Lovable, Bolt, and v0 by Vercel all produce web apps. FlutterFlow generates Flutter code, which is cross-platform but not native Swift. Draftbit is similar. Nobody else is generating production SwiftUI from natural language, because nobody else has the team that could pull it off.

The Verdict

I think Bitrig is the most compelling vibe coding product I have seen, because it is doing something genuinely new rather than applying AI to an already-solved workflow. Web apps already had low-code and no-code tools before AI showed up. Native iOS development had nothing. The gap between “I have an app idea” and “I have an app in the App Store” was months of learning and thousands of dollars in development costs. Bitrig collapses that gap to hours.

The founding team is the moat. You cannot recruit your way to this level of SwiftUI expertise. These three people helped create the framework. They understand its idioms, its edge cases, its undocumented behaviors in a way that no training data can replicate. The code Bitrig generates is not just functional, it is idiomatic, because it was built by the people who defined what idiomatic SwiftUI looks like.

The risk is market size perception. Investors might look at “iPhone app builder” and see a niche. But the number of people who want to build an iPhone app and cannot is orders of magnitude larger than the number who currently do. Bitrig is not competing for existing iOS developers. It is creating a new category of app creators who were previously locked out.

Thirty days, I want to see the App Store acceptance rate for Bitrig-generated apps. If reviewers are approving them without issues, the code quality claim is validated. Sixty days, I want to see retention: are people building one app and leaving, or are they building multiple? Ninety days, the question is whether Bitrig can handle complex apps with backends, authentication, and real data, or whether it plateaus at simple single-screen utilities. If it can scale in complexity, this company redefines who gets to build software. And the people redefining it are the same people who built the platform in the first place.