React Native engineer behavioural interview questions

As asked

Tell me about a bug that manifested on Android but not iOS. How did you isolate the root cause, and how did you fix it without breaking the iOS behavior?

Sample answer outline

The answer should describe a real cross-platform difference (e.g., Android text rendering, elevation/shadow differences, different timing of lifecycle events, keyboard behavior, back button handling). The candidate should show a systematic isolation approach (binary search through commits, platform.OS branch testing, native log inspection). The fix should be platform-safe (using Platform.select or a platform file), and the answer should mention regression testing on both platforms.

Expect these follow-ups

• How do you set up your CI to catch Android-only regressions automatically?
• What is the most surprising Android vs iOS difference you have encountered?

As asked

Tell me about a time you discovered a performance regression in a React Native app that was introduced by a colleague's PR. How did you identify it, how did you communicate it, and what was the outcome?

Sample answer outline

A strong answer uses the STAR structure: describes a specific measurable regression (frame drop, startup time increase, memory spike), explains the investigation method (profiler, git bisect, PR diff review), shows how they communicated findings respectfully and with data rather than blame, and covers what was fixed and whether a performance CI check was added to prevent recurrence. Bonus marks for mentioning automated perf budgets or baseline comparisons in CI.

Expect these follow-ups

• What automated checks would you add to catch this class of regression in the future?
• How do you handle the situation if the colleague disagrees that there is a regression?

As asked

Describe a time you had to write a native module in Swift or Kotlin for a feature with no existing library. How did you approach learning the native API, and how did you design the JS interface?

Sample answer outline

The candidate should describe a specific module (camera, BLE, payment SDK, etc.), explain how they read the native platform docs and sample code, how they defined the JS interface before writing native code (contract-first approach), and how they tested it in isolation. A strong answer mentions writing a JS mock first so the React components could be built in parallel, and handling the async nature of native callbacks via Promises or EventEmitters.

Expect these follow-ups

• How did you document the module for other team members?
• Did you publish it as an open-source library, and if not, why not?

As asked

Tell me about a major React Native version upgrade you led or were heavily involved in. What broke, how did you manage third-party library incompatibilities, and how did you coordinate the work across the team?

Sample answer outline

A strong answer covers the scope of the upgrade (e.g., RN 0.68 to 0.73), the upfront audit of breaking changes (using the RN upgrade helper), handling libraries stuck on old native APIs or old architecture, coordination strategy (feature freeze, dedicated upgrade branch, incremental merges), testing strategy (Detox, manual regression), and rollout (canary release or staged rollout). Bonus points for mentioning automated upgrade tools and having a rollback plan.

Expect these follow-ups

• How did you handle a critical library that had no upgrade path and was unmaintained?
• What would you do differently next time to reduce upgrade pain?

As asked

Describe a situation where a product or design decision was going to cause significant performance or UX problems in the React Native app. How did you raise it, and what happened?

Sample answer outline

A strong answer shows the candidate measured the impact (profiler data, user metrics) rather than guessing, raised it with specific data in a product discussion, proposed alternatives with trade-offs, and either influenced the decision or accepted it gracefully when overruled while documenting their concern. The answer should show assertiveness without being obstructionist.

Expect these follow-ups

• If you were overruled, did you still implement the feature to the best of your ability?
• How did the product outcome turn out in the end?

As asked

You joined a team that has no RN-specific code review standards and PRs are merging with obvious performance antipatterns. How do you introduce standards without slowing the team down or creating a gatekeeping culture?

Sample answer outline

The candidate should describe a collaborative approach: bringing the team together to agree on a small set of high-impact rules (no anonymous functions in render, always memoize list renderItem, no synchronous storage on the main render), documenting them in a CONTRIBUTING guide, adding ESLint rules where possible to automate enforcement, and framing code review feedback as learning rather than critique. A strong answer mentions gradually introducing these rather than a big-bang mandate.

Expect these follow-ups

• What ESLint plugins would you add for RN-specific rules?
• How do you handle a senior engineer who ignores the standards?

As asked

Tell me about a time you mentored a junior or mid-level developer who was making decisions that crossed the JS bridge unnecessarily and was causing performance issues. How did you approach the coaching?

Sample answer outline

The candidate should describe a specific situation where they identified the issue (profiling, code review), chose to teach rather than just fix, used a concrete example to explain bridge overhead, paired on the refactor, and checked in afterward to see if the mentee applied the learning independently. A strong answer shows empathy for the learning curve and avoids making the junior feel bad for the mistake.

Expect these follow-ups

• How do you balance giving a developer enough autonomy to learn from mistakes versus protecting production?
• What resources or reading did you recommend to help them go deeper?

As asked

Tell me about a time you had to ship a feature that required significantly different implementations on iOS and Android, with a tight deadline. How did you scope, parallelize, and prioritize?

Sample answer outline

The answer should describe the feature (e.g., biometrics, push, camera, payment), the specific iOS and Android divergence, how the work was split (one engineer per platform, or the candidate owning both), how the shared JS interface was agreed upfront to allow parallel work, and how testing was managed. A strong answer notes what was deferred or simplified to hit the deadline and how that technical debt was tracked.

Expect these follow-ups

• What did you cut to hit the deadline, and do you regret it?
• How did you ensure the two native implementations behaved consistently from the JS side?

As asked

One hour after releasing a new version of your React Native app, your crash rate doubles. Walk me through your incident response: how you diagnose, communicate, and decide whether to roll back or push a hotfix.

Sample answer outline

The candidate should describe checking Sentry or Firebase Crashlytics for the top crash, symbolicating the stack trace, correlating the crash with the specific commit or feature in the release, communicating to stakeholders with current severity and ETA, and making the rollback vs hotfix decision based on crash severity, the complexity of the fix, and whether CodePush can deliver it. A strong answer mentions halting the phased rollout on Google Play if the crash is detected early, and using TestFlight external groups for pre-release crash soak.

Expect these follow-ups

• How do you prevent this class of issue in future releases?
• How do you decide between a CodePush hotfix and a full store release for a critical bug?

As asked

Tell me about a time a user action in offline mode caused data loss or corruption in production. What was the root cause, how did you fix it, and what did you put in place to prevent similar issues?

Sample answer outline

The answer should describe a specific incident (double mutation on reconnect, conflicting edits, failed migration of queued items). The root cause analysis should cover the failure mode clearly. The fix should address the root cause (idempotency keys, conflict resolution policy, migration validation). Prevention measures might include chaos testing for offline scenarios, automated tests that simulate connectivity drops, and a canary release to catch similar issues early.

Expect these follow-ups

• How do you reproduce offline edge cases reliably in a test environment?
• How do you communicate data loss to affected users?

As asked

Your app was rejected by Apple 48 hours before a critical launch. Walk me through how you handled the rejection, the communication to stakeholders, and how you resolved it.

Sample answer outline

The candidate should describe a real or plausible scenario (privacy policy missing, missing purpose strings for permissions, in-app purchase circumvention). They should show calm under pressure, a fast triage to understand the exact rejection reason, a fix plan with time estimate, honest communication to stakeholders about the delay, and a post-mortem to add an App Store review checklist to the release process. Bonus for mentioning Apple's Expedited Review option for critical fixes.

Expect these follow-ups

• How do you now prevent the same category of rejection before submission?
• What is in your pre-submission checklist now?

As asked

You inherited a React Native codebase with 200 class components, no hooks, and global state in a single Redux store for everything. How did you approach modernizing it while keeping the app shipping new features?

Sample answer outline

The candidate should describe a strangler fig approach: no big-bang rewrite, instead refactoring components to hooks as they are touched for new features, introducing React Query for server state incrementally, and reducing Redux scope over time. They should show how they got team buy-in, tracked progress, and maintained a consistent codebase during the transition. A strong answer mentions lint rules to discourage new class components, pair programming to spread knowledge, and measuring impact (bundle size, test coverage, bug rate).

Expect these follow-ups

• How did you handle a class component that used lifecycle methods with complex state that had no clean hooks equivalent?
• What metrics did you use to know the refactor was paying off?