Migrating Coinbase's 56 Million Users to React Native: Key Lessons and Takeaways

Introducing Harry Tori and Nick Cherry, discussing their backgrounds and experiences with database and application performance, focusing on their work at Coinbase and Forecaster.

Nick's transition from a backend Ruby Dev to working with React Native at Coinbase, facing challenges and issues with mobile clients, leading to the decision to replatform the whole mobile app to React Native.

Exploring the performance issues faced during the React Native rewrite at Coinbase, including issues with rendering, crashes, and low velocity, leading to the adoption of instrumentation tools.

Discussing the process of identifying performance problems in React Native, focusing on issues with rendering counts, FPS drops, and the impact on user experience.

Exploring strategies to optimize React Native performance, including memoization, context-based rendering, and navigation challenges on mobile compared to web.

Discussing navigation challenges in React Native, comparing the differences between mobile and web navigation, and the impact on performance and rendering.

Exploring solutions to navigation issues in React Native, including delegated keys, managing data layers, and optimizing performance for both Android and iOS platforms.

Nick's experience in building the FarCaster mobile app, utilizing libraries like React Navigation and Tailwind CSS, as well as discussing performance challenges and solutions during the development process.

Addressing the use of native libraries in the FarCaster app development, overcoming performance issues with intensive operations by using a native library and bridging it with React Native.

Highlighting the benefits of Expo in app development, including Expo modules, CI pipelines, OTA updates, and the ease of development for React Native apps, particularly in larger teams and enterprises.

Discusses the importance of quickly fixing small issues in business logic in crypto and financial apps to mitigate problems affecting large numbers of users.

Considers the possibility of starting a mobile application again in 2024 and the focus on elegant data normalization patterns and skepticism towards server-side solutions like GraphQL.

Focuses on the challenges of synchronizing data in apps, the need for unique IDs for data sources, and the benefits of creating a pattern for synchronization.

Discusses a prototype for reducing network latency by batching requests in mobile applications, which showed a significant reduction in network latency.

Explores how AI can assist in frontend development, specifically in optimizing rendering, identifying data access patterns, and prefetching data for improved performance.

Highlights the importance of analyzing system performance holistically, especially in high-traffic scenarios like price notifications in crypto, and the potential benefits of anomaly detection and AI in proactive problem-solving.