How do you derisk achieving goals in large projects with React Native?

The product vision team set a 200-millisecond app launch target for a consumer electronics project. The team wanted to use JavaScript and existing web talent, but ruled out a browser-based interface because the previous product had serious performance and stability problems.

React Native let engineers keep working in JavaScript while using a declarative UI approach. By this project, major apps had shipped with React Native, giving the team stronger evidence that the technology could work repeatedly than was available during an earlier project at BlueJeans.

Before designs were ready, the team recreated existing Ember-based WebKit apps in React Native. This let them check that the needed controls worked and look for hidden issues such as memory leaks, while trying to reuse the QA team's UI automation.

The initial group had six to eight engineers, trained in small batches by practitioners from Callstack and This Dot Labs. Training emphasized unit testing, performance pitfalls that cause unnecessary re-renders, and using React DevTools to find them, especially on lower-end hardware.

After about six months, the team added React Navigation to its training and contributed pull requests. The guest connected this ecosystem approach to faster onboarding: new hires could make useful contributions that passed review and stayed in the codebase.