Mastering Kotlin Coroutines: Complete 2026 Guide
Learn to master Kotlin coroutines for Android development: suspend functions, scopes, dispatchers, and advanced patterns.

Kotlin coroutines have revolutionized asynchronous programming on Android. Gone are the days of callback hell and deprecated AsyncTask: with coroutines, you write asynchronous code that looks synchronous while remaining performant and maintainable.
Coroutines are lightweight (thousands can run on a single thread), natively cancellable, and integrate seamlessly with Jetpack and the modern Android ecosystem.
Understanding the Fundamentals
Before diving into code, let's understand what makes coroutines so powerful.
What is a Coroutine?
A coroutine is an instance of suspendable computation. Unlike threads, coroutines don't block: they suspend their execution and free the thread for other tasks.
import kotlinx.coroutines.*
fun main() = runBlocking {
launch {
delay(1000L) // Suspend without blocking
println("World!")
}
println("Hello,")
}
// Output: Hello, World!Suspend Functions: The Heart of Coroutines
The suspend keyword indicates that a function can suspend coroutine execution without blocking the thread.
suspend fun fetchUserData(userId: String): User {
return withContext(Dispatchers.IO) {
// Network call - runs on an IO thread
apiService.getUser(userId)
}
}
suspend fun fetchUserWithPosts(userId: String): UserWithPosts {
// Sequential execution
val user = fetchUserData(userId)
val posts = fetchUserPosts(userId)
return UserWithPosts(user, posts)
}Golden rule: A suspend function can only be called from another suspend function or from within a coroutine.
Coroutine Scopes
Scope defines the lifecycle of your coroutines. This is crucial for avoiding memory leaks.
viewModelScope: The Scope for ViewModels
class UserViewModel(
private val userRepository: UserRepository
) : ViewModel() {
private val _uiState = MutableStateFlow<UserUiState>(UserUiState.Loading)
val uiState: StateFlow<UserUiState> = _uiState.asStateFlow()
fun loadUser(userId: String) {
viewModelScope.launch {
_uiState.value = UserUiState.Loading
try {
val user = userRepository.getUser(userId)
_uiState.value = UserUiState.Success(user)
} catch (e: Exception) {
_uiState.value = UserUiState.Error(e.message)
}
}
}
}
sealed class UserUiState {
object Loading : UserUiState()
data class Success(val user: User) : UserUiState()
data class Error(val message: String?) : UserUiState()
}lifecycleScope: For Activities and Fragments
class UserFragment : Fragment() {
private val viewModel: UserViewModel by viewModels()
override fun onViewCreated(view: View, savedInstanceState: Bundle?) {
super.onViewCreated(view, savedInstanceState)
viewLifecycleOwner.lifecycleScope.launch {
viewLifecycleOwner.repeatOnLifecycle(Lifecycle.State.STARTED) {
viewModel.uiState.collect { state ->
when (state) {
is UserUiState.Loading -> showLoading()
is UserUiState.Success -> showUser(state.user)
is UserUiState.Error -> showError(state.message)
}
}
}
}
}
}Never use GlobalScope in an Android application. Coroutines launched with GlobalScope aren't tied to any lifecycle and can cause memory leaks.
Dispatchers: Controlling Execution
Dispatchers determine which thread your coroutine runs on.
The 4 Main Dispatchers
// Main: main thread (UI)
viewModelScope.launch(Dispatchers.Main) {
textView.text = "UI update"
}
// IO: I/O operations (network, database)
viewModelScope.launch(Dispatchers.IO) {
val data = repository.fetchFromNetwork()
}
// Default: CPU-intensive computations
viewModelScope.launch(Dispatchers.Default) {
val result = heavyComputation(data)
}
// Unconfined: inherits caller context (rare use)withContext: Switching Dispatchers
class ImageProcessor {
suspend fun processImage(bitmap: Bitmap): Bitmap {
return withContext(Dispatchers.Default) {
// CPU-intensive processing on Default
applyFilters(bitmap)
}
}
suspend fun saveToGallery(bitmap: Bitmap) {
withContext(Dispatchers.IO) {
// Disk write on IO
saveToFile(bitmap)
}
}
}
// Usage in ViewModel
viewModelScope.launch {
val processed = imageProcessor.processImage(originalBitmap)
imageProcessor.saveToGallery(processed)
// Automatic return to Main for UI update
_uiState.value = UiState.Success(processed)
}Parallel Execution with async/await
To execute tasks in parallel and combine their results, use async.
suspend fun loadDashboard(): Dashboard {
return coroutineScope {
// Parallel launch
val userDeferred = async { userRepository.getUser() }
val statsDeferred = async { statsRepository.getStats() }
val notificationsDeferred = async { notificationRepository.getNotifications() }
// Await results
Dashboard(
user = userDeferred.await(),
stats = statsDeferred.await(),
notifications = notificationsDeferred.await()
)
}
}With async, all 3 calls execute in parallel. If each call takes 1 second, total time is ~1 second instead of 3 seconds sequentially.
Error Handling
Classic try/catch
viewModelScope.launch {
try {
val user = userRepository.getUser(userId)
_uiState.value = UiState.Success(user)
} catch (e: HttpException) {
_uiState.value = UiState.Error("Server error: ${e.code()}")
} catch (e: IOException) {
_uiState.value = UiState.Error("Network error")
} catch (e: Exception) {
_uiState.value = UiState.Error("Unexpected error")
}
}CoroutineExceptionHandler
class UserViewModel : ViewModel() {
private val exceptionHandler = CoroutineExceptionHandler { _, throwable ->
_uiState.value = UiState.Error(throwable.message)
Timber.e(throwable, "Error in coroutine")
}
fun loadUser(userId: String) {
viewModelScope.launch(exceptionHandler) {
val user = userRepository.getUser(userId)
_uiState.value = UiState.Success(user)
}
}
}Result Wrapper Pattern
sealed class Result<out T> {
data class Success<T>(val data: T) : Result<T>()
data class Error(val exception: Throwable) : Result<Nothing>()
}
suspend fun <T> safeApiCall(apiCall: suspend () -> T): Result<T> {
return try {
Result.Success(apiCall())
} catch (e: Exception) {
Result.Error(e)
}
}
// Usage
class UserRepository(private val api: UserApi) {
suspend fun getUser(id: String): Result<User> = safeApiCall {
api.getUser(id)
}
}
// In ViewModel
viewModelScope.launch {
when (val result = userRepository.getUser(userId)) {
is Result.Success -> _uiState.value = UiState.Success(result.data)
is Result.Error -> _uiState.value = UiState.Error(result.exception.message)
}
}Ready to ace your Android interviews?
Practice with our interactive simulators, flashcards, and technical tests.
Cancellation: Cleaning Up Properly
Coroutines support cooperative cancellation. This is essential for avoiding resource leaks.
Automatic Cancellation with Scopes
class SearchViewModel : ViewModel() {
private var searchJob: Job? = null
fun search(query: String) {
// Cancel previous search
searchJob?.cancel()
searchJob = viewModelScope.launch {
delay(300) // Debounce
val results = searchRepository.search(query)
_searchResults.value = results
}
}
}Checking for Cancellation
suspend fun processLargeList(items: List<Item>) {
items.forEach { item ->
// Check if coroutine is cancelled
ensureActive()
processItem(item)
}
}
suspend fun downloadFiles(urls: List<String>) = coroutineScope {
urls.map { url ->
async {
try {
downloadFile(url)
} catch (e: CancellationException) {
cleanupPartialDownload(url)
throw e // Re-throw to propagate cancellation
}
}
}.awaitAll()
}Never swallow CancellationException: if you catch Exception, re-throw CancellationException so cancellation propagates correctly.
Flow: Reactive Programming
Flow is the coroutines equivalent of RxJava Observable, but simpler and integrated.
Creating and Collecting a Flow
fun getUsers(): Flow<List<User>> = flow {
while (true) {
val users = userApi.getUsers()
emit(users)
delay(5000) // Poll every 5 seconds
}
}
// Flow from Room
@Dao
interface UserDao {
@Query("SELECT * FROM users")
fun getAllUsers(): Flow<List<User>>
}
// Collecting in ViewModel
viewModelScope.launch {
userDao.getAllUsers()
.catch { e -> _uiState.value = UiState.Error(e.message) }
.collect { users ->
_uiState.value = UiState.Success(users)
}
}StateFlow vs SharedFlow
class EventViewModel : ViewModel() {
// StateFlow: keeps last value, ideal for UI state
private val _uiState = MutableStateFlow(UiState.Initial)
val uiState: StateFlow<UiState> = _uiState.asStateFlow()
// SharedFlow: for one-shot events (navigation, snackbar)
private val _events = MutableSharedFlow<UiEvent>()
val events: SharedFlow<UiEvent> = _events.asSharedFlow()
fun onButtonClick() {
viewModelScope.launch {
_events.emit(UiEvent.NavigateToDetail)
}
}
}
sealed class UiEvent {
object NavigateToDetail : UiEvent()
data class ShowSnackbar(val message: String) : UiEvent()
}Essential Flow Operators
userRepository.getUsers()
.map { users -> users.filter { it.isActive } }
.distinctUntilChanged()
.debounce(300)
.flatMapLatest { users ->
fetchUserDetails(users)
}
.catch { e ->
emit(emptyList())
}
.onEach { users ->
analytics.logUserCount(users.size)
}
.stateIn(
scope = viewModelScope,
started = SharingStarted.WhileSubscribed(5000),
initialValue = emptyList()
)Advanced Patterns
Retry with Exponential Backoff
suspend fun <T> retryWithBackoff(
times: Int = 3,
initialDelay: Long = 100,
maxDelay: Long = 1000,
factor: Double = 2.0,
block: suspend () -> T
): T {
var currentDelay = initialDelay
repeat(times - 1) { attempt ->
try {
return block()
} catch (e: Exception) {
Timber.w("Attempt ${attempt + 1} failed, retrying in ${currentDelay}ms")
}
delay(currentDelay)
currentDelay = (currentDelay * factor).toLong().coerceAtMost(maxDelay)
}
return block()
}
// Usage
val user = retryWithBackoff {
userApi.getUser(userId)
}Timeout
suspend fun fetchWithTimeout() {
try {
val result = withTimeout(5000L) {
api.fetchData()
}
processResult(result)
} catch (e: TimeoutCancellationException) {
showError("Request took too long")
}
}
// Or with a default value
val result = withTimeoutOrNull(5000L) {
api.fetchData()
} ?: defaultValueConclusion
Kotlin coroutines have become essential for modern Android development. They offer an elegant and performant approach to asynchronous programming, perfectly integrated with the Jetpack ecosystem.
Checklist
- ✅ Use
viewModelScopeandlifecycleScopeto avoid leaks - ✅ Choose the right Dispatcher (Main, IO, Default)
- ✅ Handle errors with try/catch or Result wrapper
- ✅ Implement cooperative cancellation
- ✅ Use Flow for reactive data streams
- ✅ Prefer StateFlow for UI state, SharedFlow for events
Start practicing!
Test your knowledge with our interview simulators and technical tests.
Mastering coroutines will give you a significant advantage in your Android projects and technical interviews. Practice regularly and don't hesitate to explore advanced patterns!
Can you spot the bug in Android?
One real snippet, one hidden bug, one attempt a day. No account needed to try.

Written by
Anthony Fillion-MailletFounder of SharpSkill
Full-stack developer for over 10 years. Runs SharpSkill and answers for everything published here.
Updated on March 28, 2026
Tags
Share
Related articles

Kotlin Flow vs StateFlow vs SharedFlow: Android Interview Questions in 2026
The Kotlin Flow vs StateFlow vs SharedFlow questions Android interviewers ask in 2026, with clear answers, a comparison table, and production-ready code.

Room Database in Android 2026: Migrations, Relations and Coroutines
Master Room database in Android with migrations, entity relations, Kotlin coroutines and Flow. Complete tutorial with production-ready patterns for 2026.

Kotlin 2.3 for Android: Name-Based Destructuring, KMP and Interview Questions 2026
Kotlin 2.3 interview questions covering name-based destructuring, Kotlin Multiplatform, context parameters, coroutines and Flow. Prepare for Android developer interviews in 2026 with real-world code examples.