Mike Grabowski

Writing code is only one part of delivering software. An agentic development loop begins with the intent behind a change and continues until its impact can be observed in production.

That means understanding the product and the existing system, deciding what should change, implementing it across shared and platform-specific layers, verifying the result in the running application, releasing it safely, and learning from how the software behaves after deployment. Each step should produce context for the next one.

Cross-platform products make this especially difficult. A change that looks correct in the repository can still fail at the boundary between JavaScript and native code, behave differently across platforms, or create problems that only become visible after release.

This talk examines what it takes to close the full ADLC loop across mobile, web, and native. Drawing on Callstack's work in React Native and AI-native engineering, it maps the development system required to carry intent, context, and evidence through the entire lifecycle and turn agent output into dependable product outcomes.

Apple's first on-device foundation model gave developers a new execution path for AI features. iOS 27 expands what the system can do, while Android is moving in the same direction. If you last looked at this space around the initial release, the trade-offs and possible products have already changed.

System-provided models have a distinctive place in the stack. They ship with the device, are optimized for its hardware, work offline, keep data local, and remove the need to package and maintain a model yourself. Their capabilities are defined by the platform and differ from cloud inference and models supplied by the application.

This talk reviews the additions announced at WWDC and our work to bring the same approach to React Native on Android. Our integrations are built around the Vercel AI SDK, so applications can use familiar APIs regardless of where inference runs. We will look at what system models can now do, where their constraints appear, and which product features are particularly well suited to them.

If your last experiment stopped at the initial Apple Intelligence release, this talk will give you a reason and a path to try again.

AI coding agents can generate React Native code quickly, but they still need a reliable way to see, touch, debug, and verify the app they are changing. This talk introduces Agent Device, Callstack’s agent-native automation layer for mobile apps, and shows how it gives coding agents a practical feedback loop on iOS, Android, TV, and desktop targets.

We’ll look at what that means in practice: how agents inspect the UI, interact with the app, collect evidence, profile React Native behavior, and turn exploratory work into replayable checks. Then we’ll show practical workflows for React Native teams: letting an agent implement and verify a feature, debugging a broken screen, turning exploratory QA into deterministic checks, and supporting AI-assisted migrations where implementation and testing feed into each other.

The goal is to show how agent-assisted coding changes when the agent can work against the real app, not just the source code.

Module Federation lets you organize code into federated modules, dynamically loaded at runtime within a host application—a core mechanism behind Super Apps. On the web, this is typically handled with Webpack.Now, with Re.Pack as a Metro alternative, React Native supports it too. This allows existing native Super Apps to transition to React Native and enables truly cross-platform modules, letting developers use React and JavaScript without concern for where modules run.In this talk, we’ll explore these possibilities, best practices and real-world examples to help you build efficient, cross-platform Super Apps.