Intermediate interview questions in Mobile Dev, page 7
Axios interceptors for auth headers
Interceptors are hooks that run before requests or after responses, centralizing concerns like auth, logging, and token refresh.
Which side-effect handler for a one-time coroutine on composition entry?
Tests Compose side-effect API boundaries. Answer: LaunchedEffect(Unit); it runs once and auto-cancels on exit. Contrast with rememberCoroutineScope for user-triggered callbacks. Red flag: picking SideEffect or rememberCoroutineScope and missing cancellation.
Which Compose side-effect for a one-time coroutine action?
Tests if you can choose the right Compose side-effect for a one-time coroutine. A great answer names LaunchedEffect(Unit), explains it runs once and cancels on exit. A red flag is suggesting rememberCoroutineScope directly in the composable body.
Which side-effect handler for a one-time coroutine action in Compose?
This tests your understanding of Compose lifecycles and side-effect handlers. A good answer names LaunchedEffect(Unit), explaining the constant key ensures the effect runs only once. A red flag is suggesting launching coroutines on every recomposition.
Cancelling fetch with AbortController
Pass an AbortController signal to fetch and call abort() on unmount to avoid wasted work, race conditions, and state updates on gone components.
How do you display a large list efficiently in Jetpack Compose?
This tests understanding of Compose lazy versus eager layout. A strong answer names LazyColumn or LazyRow, notes visible-only composition, and contrasts Column which lays out everything upfront.
How to implement efficient large lists in Jetpack Compose?
This tests your grasp of Compose performance. Use LazyColumn to compose only visible items, unlike a Column which renders all items at once. A red flag is suggesting a scrollable Column, which has severe performance costs for large lists.
Implement an efficient list in Jetpack Compose
Tests your grasp of UI virtualization in Compose. Answer: Use LazyColumn as it only composes visible items. Contrast this with Column, which composes all items at once, causing poor performance. Mention the items DSL.
GraphQL versus REST data fetching
One endpoint, client-specified fields avoid over- and under-fetching; Apollo's useQuery returns loading, error, and data while caching results.
Why store auth tokens in Keychain, not UserDefaults?
UserDefaults is an unencrypted plist readable from backups and on jailbroken devices; use Keychain Services, which stores encrypted items with access control; save with SecItemAdd and read with…
Describe a pattern for background Core Data fetches and UI updates
This tests NSManagedObjectContext concurrency and queue confinement. Strong answer: create a background context, fetch, pass objectIDs or structs to main, then main context fetches by ID to update UI. Red flag: passing NSManagedObjects across threads.
Explain the core responsibilities of RouterDelegate, RouteInformationParser, and Router in Flutter
Tests Navigator 2.0 architecture split. RouteInformationParser converts URLs to typed config; RouterDelegate owns state and builds the Navigator; Router wires them together and handles engine intents. Red flag: conflating parsing and state or omitting Router.
Describe lightweight vs heavyweight Core Data migration and a heavyweight example
This tests inferred versus manual Core Data migrations. Contrast lightweight additive changes with heavyweight transforms like entity splits, citing promoting a string to a new entity. A red flag is claiming renames always need custom mapping.
Implement a GoRouter path parameter route and access userId in a widget
This tests GoRouter declarative routing and GoRouterState. Define a GoRoute with path '/users/:userId', read state.params['userId'] in the builder, and handle null-safety. A red flag is parsing the URI manually or using ModalRoute.of instead of state.params.
Per-tab navigation stacks with StatefulShellRoute
StatefulShellRoute keeps a separate Navigator and preserved state per branch, with a shared shell scaffold.
Why is LiveData<Boolean> problematic for one-time ViewModel events?
State is not events; suggest an Event wrapper with a consumed flag, or SharedFlow or Channel with no replay.
Why is LiveData<Boolean> bad for one-time ViewModel events?
Tests if you know LiveData holds state, not events. A good answer explains that LiveData re-emits on configuration changes, causing repeated events. The solution is to model events as part of the UI state and consume them.
Why is `LiveData<Boolean>` bad for one-time ViewModel events?
This tests your grasp of state vs. events. A LiveData<Boolean> is state; it re-triggers on rotation. Instead, expose events via a StateFlow and have the UI call an eventConsumed() function on the ViewModel to maintain unidirectional data flow.
Platform files versus Platform.OS checks
Use platform files for large divergence and platform-only deps, use Platform.OS for small inline tweaks.
How would you use a nested navigation graph for a login flow?
Tests modular navigation scoping. Strong answers group the three screens under a nested graph with its own start destination, noting encapsulation and reusability.
We are hiring for this. Every open role lists the topics its interview covers, so you can prepare for the real thing rather than guessing.
See open roles