Atos Android Developer Interview Preparation

The Activity lifecycle is not a durable data boundary. Separate UI state from the component instance and keep a defined source of truth in the data layer. Distinguish configuration changes from process death; state that must survive longer needs an appropriate saved or persistent representation.

Check your understanding: Rotate the device and separately recreate the process, checking the intended state.

Common trap: Assuming a ViewModel is durable storage across process death.

Concept reference: Android: app architecture

Expose state from a data layer with an explicit source of truth, freshness and error model. Decide which actions can be queued and how they reconcile later. Keep network and storage details out of the UI component, and test a reconnect that returns older or conflicting data.

Check your understanding: Disconnect during an edit and verify the documented reconciliation behavior.

Common trap: Replacing useful cached data with a generic error screen.

Concept reference: Android: app architecture

A scope connects task lifetime, cancellation and failure propagation. Work started without a deliberate owner can outlive the screen or operation that needs it. Choose an appropriate lifecycle or application boundary and ensure cancellation does not leave half-applied local state.

Check your understanding: Cancel the owner and verify child work finishes or cancels as designed.

Common trap: Launching every task in an unrestricted global scope.

Concept reference: Kotlin: coroutine fundamentals

First, identify technical constraints and define measurable service objectives. Next, trace a user action through lifecycle, local state, network and restored UI. Contrast architectural trade-offs across simplicity, correctness, maintainability and scale, explicitly mitigate the risk of a partially deployed reader cannot understand the new representation, and confirm system stability using a compatibility contract, expand-and-contract rollout and rollback rehearsal.

Common trap: Reaching for a specific library or framework before defining constraints, failure envelopes, and automated verification criteria.

Prioritise the failure mode exhibiting the highest user blast radius and lowest observability. Formulate an explicit containment boundary, implement idempotent retries with jitter, and establish an automated rollback threshold. Verify resilience through a compatibility test matrix and a staged rollout metric.

Common trap: Relying on passive monitoring dashboards without defining explicit error-budget alerts, rollback triggers, and verified recovery procedures.

Structure the response using Context-Decision-Tradeoff-Result: articulate the business and technical constraints, compare viable alternatives, explain the implementation (trace a user action through lifecycle, local state, network and restored UI), and document the accepted trade-off. Provide concrete proof: a project example, measured result and repeatable verification step.

Common trap: Speaking only in high-level abstractions or team accomplishments without detailing your direct implementation decisions, trade-offs, and measured results.

Formulate a falsifiable hypothesis from observable telemetry before altering configurations. Then compare lifecycle events, persisted state, network retries and device-specific traces. Isolate the defect to the smallest reproducible boundary, validate root cause with evidence, and confirm full resolution using queue-depth metrics, bounded retry behaviour and a recovery drill.

Common trap: Applying speculative fixes or restarting services blindly without establishing an observable signal connected to a falsifiable hypothesis.

Produce a constrained-device test matrix with offline recovery evidence. Document baseline assumptions, technical mechanism (trace a user action through lifecycle, local state, network and restored UI), rejected alternatives, bounded failure envelopes, and deterministic pass criteria. Supply reproducible verification via a concurrency test and an audit trail demonstrating conflict handling.

Common trap: Presenting architecture diagrams or slides lacking automated unit/integration tests, observable metrics, or automated rollback configurations.