JPJobPrepfull-stack interview
RoadmapsJS CompilerStar on GitHub

Career roadmap

Android Engineer

Kotlin, Compose and the Android platform — build for the widest and most fragmented device base in the world.

Time
7-10 months part-time
Entry bar
Programming experience in any language. Any laptop that runs Android Studio.
Stages
5 · 25 topics
0/25 studied0%

Before you start Android

  • Object-oriented programming fundamentals
  • Android Studio installed with an emulator running
  • Basic Git

Kotlin and the platform

5-6 weeks · 0/5 topics

Kotlin idioms and the Android lifecycle — the two things every interview opens with.

  1. Null safety, data classes and scope functions are assumed knowledge, not bonus points.

    • Null safety and platform types
    • Data classes, sealed classes, enums
    • Extension functions and scope functions
    • Collections and sequences
  2. Structured concurrency is the single most-asked Android topic in current interviews.

    • Suspend functions and coroutine scopes
    • Dispatchers and thread confinement
    • Flow, StateFlow and SharedFlow
    • Cancellation and exception handling
  3. Rotation and process death are the classic Android trap questions.

    • Activity and fragment lifecycle
    • ViewModel and saved state
    • Process death and restoration
    • Configuration changes and resources
  4. Components, intents and permissions — the vocabulary of the platform.

    • Activities, services, broadcast receivers
    • Intents, intent filters and deep links
    • Runtime permissions
    • Manifest, resources and qualifiers
  5. Build variants and dependency management, plus why the build takes four minutes.

    • Gradle Kotlin DSL and version catalogs
    • Build types, flavours and variants
    • R8 shrinking and ProGuard rules
    • Build speed and modularisation

BuildA two-screen app that survives rotation and process death with state intact.

Jetpack Compose

5-7 weeks · 0/5 topics

Compose is the default for new Android work. Recomposition is the interview topic.

  1. Declarative UI on Android, and the recomposition model that decides performance.

    • Composables, state and recomposition
    • remember, derivedStateOf, rememberSaveable
    • Side effects: LaunchedEffect, DisposableEffect
    • Stability and skipping
  2. Material 3, dynamic colour and custom layouts.

    • Row, Column, Box and modifiers order
    • LazyColumn and item keys
    • Material 3 and dynamic colour
    • Custom layouts and measurement
  3. Type-safe navigation and deep links, plus back stack behaviour.

    • Navigation Compose and typed routes
    • Back stack and up navigation
    • Deep links and app links
    • Nested graphs and shared state
  4. Most existing Android codebases are XML views. You will maintain them.

    • XML layouts and ConstraintLayout
    • RecyclerView and DiffUtil
    • View binding
    • Interop between Views and Compose
  5. Foldables, tablets and TalkBack. Increasingly part of the Play Store quality bar.

    • TalkBack semantics in Compose
    • Window size classes and adaptive layouts
    • Font scaling and contrast
    • Foldable and large screen support

BuildA Compose app with lists, navigation, theming and a custom layout, supporting dark mode.

Architecture and data

5-6 weeks · 0/5 topics

Google's recommended architecture is the expected answer; knowing its costs is the senior one.

  1. UI layer, domain layer, data layer — and unidirectional data flow throughout.

    • MVVM and MVI in practice
    • Repository pattern and data sources
    • Use cases and the domain layer
    • Single source of truth
  2. Hilt is the standard. Being able to explain scoping is the actual test.

    • Hilt modules, components and scopes
    • Constructor injection and testability
    • Manual DI and when it is enough
    • Koin as an alternative
  3. Local database plus migrations, which is where production Android apps break.

    • Room entities, DAOs and relations
    • Flow-returning queries
    • Migrations and destructive fallback
    • DataStore for preferences
  4. Retrofit, OkHttp and a caching strategy that works on a train.

    • Retrofit and OkHttp interceptors
    • Serialization with kotlinx
    • Offline-first with a single source of truth
    • WorkManager for deferred sync
  5. Unit tests for the domain, instrumented tests for the UI, and Turbine for flows.

    • JUnit, MockK and Turbine
    • Compose UI testing
    • Espresso for view-based screens
    • Test doubles for repositories

BuildRefactor your app into data/domain/UI layers with a repository, offline cache and dependency injection.

Performance and release

4-5 weeks · 0/5 topics

Android fragmentation means performance work is measured on real, cheap devices.

  1. Startup time, jank and ANRs are the Play Console vitals that gate your listing.

    • Android Studio Profiler and Perfetto
    • Baseline Profiles and startup time
    • Jank, frame timing and ANRs
    • Memory leaks with LeakCanary
  2. In many markets, install size decides whether the app is installed at all.

    • App Bundles and dynamic delivery
    • R8 and resource shrinking
    • Feature modules on demand
    • Asset optimisation
  3. Tracks, staged rollout and policy compliance — real operational knowledge.

    • Internal, closed, open and production tracks
    • Staged rollout and halting a release
    • Play policy and data safety form
    • Signing and Play App Signing
  4. Automated builds, tests and Play uploads on every merge.

    • GitHub Actions for Android
    • Automated tests on emulators
    • Fastlane supply for uploads
    • Build caching
  5. Play Console vitals plus Crashlytics is how you learn what a thousand device models do.

    • Crashlytics and non-fatal reporting
    • Android vitals thresholds
    • Deobfuscation mapping files
    • Device-specific bug triage

BuildProfile startup and jank on a low-end device, fix one issue, and ship to Play with a staged rollout.

Interview preparation

4-5 weeks · 0/5 topics

Android loops test Kotlin, coroutines, lifecycle and architecture, plus a live build.

  1. Coroutine scoping, lifecycle traps and recomposition come up in nearly every loop.

    • viewModelScope vs lifecycleScope
    • StateFlow vs LiveData vs SharedFlow
    • Why did this composable recompose
    • Handling process death correctly
  2. Build a screen that fetches and displays data, with loading and error states.

    • Compose screen from scratch in 45 minutes
    • Repository and ViewModel wiring
    • Error and empty state handling
    • Talking through decisions while coding
  3. Design an offline-first app, an image loader, or a sync engine.

    • Offline-first architecture
    • Image loading and caching pipeline
    • Background sync with WorkManager
    • Modularisation for a large team
  4. Standard at product companies and most large service companies.

    DSA Coding Questions
    • Arrays, strings and hash maps
    • Trees, graphs and recursion
    • Complexity analysis
    • Kotlin idioms in solutions
  5. A live Play listing with a clean modular repo is the strongest possible evidence.

    • Published app with screenshots
    • Multi-module open-source repo
    • Baseline Profile and startup numbers
    • Accessibility support demonstrated

BuildA published Play Store app with a modular architecture and a README that explains the layering.

Android tools on your CV

  • Kotlin
  • Jetpack Compose
  • Coroutines & Flow
  • Hilt
  • Room
  • Retrofit
  • WorkManager
  • Android Studio
  • Perfetto

What Android employers ask to see

  • An app on Google Play with staged rollout used at least once
  • A multi-module repo following the recommended architecture
  • A startup-time improvement backed by Baseline Profile numbers
  • An offline-first feature with a documented sync strategy

Huge installed base and steady enterprise demand, especially in India, South-East Asia and fintech. Less crowded than web, and Kotlin skills transfer to backend work.

Content last reviewed 2026-08-31. Guidance only — no institute or paid placement is endorsed anywhere in this book.