Shopify Abandons React Native for Native Development: How AI Is Reshaping Mobile Tech Decisions

Shopify drops React Native for native Swift/Kotlin as AI tools erode cross-platform cost advantages.
Shopify has announced it is migrating its mobile apps from React Native back to native Swift and Kotlin. The 2020 switch to React Native was driven by the need to reduce dual-platform maintenance costs, but AI coding assistants can now efficiently handle cross-platform code translation, testing, and review — significantly undermining the core value proposition of cross-platform frameworks. Shopify is also making an orderly exit from the React Native open source ecosystem, handing off key libraries to the community. This marks a clear signal that AI is redefining the cost model behind technology selection.
Shopify's Technical Pivot: From Cross-Platform Back to Native
Shopify recently published a thought-provoking engineering blog post titled Native is now the future of mobile at Shopify, announcing that it is migrating its mobile applications from React Native back to separate native codebases in Swift and Kotlin. The move carries symbolic weight — Shopify was once one of the most committed enterprise advocates for the React Native ecosystem, and now it's choosing to return to native.
What you might not notice at first glance is that Shopify never disparages React Native. In this carefully worded post, they fully acknowledge the value React Native delivered over the past six years. This isn't a bitter breakup — it's a rational reassessment driven by a shift in technological realities.

Why Did Shopify Choose React Native in the First Place?
Looking back at 2020, Shopify's decision to switch from native to React Native was driven by motivations shared by nearly every team that adopts a cross-platform approach:
- Avoid duplicating work: No need to implement the same feature separately for iOS and Android
- Unify the tech stack: Let developers work across platforms without a strict divide between iOS and Android engineers
- Focus on delivering value: Spend time shipping new features rather than chasing feature parity between two platforms
These three reasons were airtight at the time. The biggest pain point of native development was the doubled maintenance cost of two codebases, along with the inevitable gaps in features and experience between iOS and Android. React Native addressed these issues with a single JavaScript/TypeScript codebase, making it a sensible choice for many large enterprises.
AI Coding Assistants Have Rewritten the Cost Equation
So what changed Shopify's mind? The answer points to the hottest variable of the past two years — AI coding assistants.
Shopify's blog offers this key insight:
Native still means building and maintaining software on two platforms — that cost hasn't gone away. What has changed is that AI Agents can now do enough of the implementation, translation, testing, and review work that it's no longer the deciding factor it was in 2020.
The weight of this statement shouldn't be underestimated. It means that the core value proposition of cross-platform frameworks — reducing the cost of maintaining two platforms — is being eroded by AI coding tools.
From "Saving Headcount" to "AI Fills the Gap": A Paradigm Shift in Development Efficiency
In the past, teams' biggest concern was "having to write the same thing twice." Now, AI Agents can shoulder much of the repetitive implementation work: "translating" a feature from one platform to the other, automatically generating test cases, and assisting with code reviews. When the marginal cost of these tasks is dramatically reduced by AI, the inherent doubled cost of native development is no longer an insurmountable barrier.
In other words, React Native originally solved a "labor cost" problem, and AI Agents now solve the same problem from a different angle — not by letting you write things once, but by making writing it a second time nearly effortless. When two paths converge on the same destination, the natural advantages of native development in performance, platform capabilities, and user experience regain the upper hand.
The Ripple Effect on the React Native Open Source Ecosystem
Shopify wasn't just a consumer of React Native — it was a major contributor to the ecosystem. It maintains three influential open source libraries:
- react-native-skia: A high-performance rendering library built on the Skia graphics engine, widely depended upon across the community
- flash-list: A high-performance list component and popular FlatList alternative
- restyle: A type-safe styling system
Shopify's handling of this transition has been relatively responsible: react-native-skia and flash-list, both widely used, are actively seeking new maintainers to ensure the community isn't left without support; restyle, which has a smaller user base, will be archived by the end of 2026.
This arrangement sends a clear signal: Shopify is making an orderly exit from the React Native ecosystem. For development teams that depend on these libraries, evaluating alternatives and migration paths sooner rather than later is a priority.
Is Cross-Platform Approaching Its Twilight?
It's worth taking a measured view here — Shopify's decision doesn't signal the end of React Native or cross-platform development broadly. It looks more like a targeted decision made by a well-resourced, top-tier company with extremely high standards for native experience, made as AI capabilities have gradually matured.
For most small and mid-sized teams, AI can equally enhance the development efficiency of cross-platform approaches, and a return to dual codebases may not be necessary. The real takeaway is this: AI is redefining the cost model behind technology decisions. Many architectural choices made around 2020 rest on cost assumptions that may have quietly become obsolete — and are worth revisiting.
Shopify's return to native development is less a simple swing of the pendulum and more a landmark case of engineering decision-making logic being restructured in the age of AI. As the cost of implementation, translation, and testing gets dramatically compressed by AI, the standard by which we measure "what makes a good architecture" needs to be updated accordingly.
Related articles

LangChain + MCP: From Core Concepts to Agent Tool Calling in Practice
Learn how LangChain and MCP work together — covering LLM tool calling, Agent architecture, and conversation history management to build real-world AI applications.

Probabilistic Machine Learning: Why It's the Cornerstone to Unlocking the ML Black Box
Without probability theory, ML is always a black box. This article explores why probabilistic foundations are essential for understanding machine learning algorithms, Bayes' theorem, MLE, and more.

Optimization Pitfalls in Self-Evolving LLM Agents: Value Concentration and Budget-Splitting Problems
HARNESSEVO research reveals 3 key LLM agent harness optimization findings: value concentrates in reflection/control slots, uniform budget splitting is harmful, and credit assignment must precede structured evolution.