Java Backend Engineer OOPS Concepts Interview Guide

Choose composition when the relationship is about using a capability rather than being substitutable for the base type. Delegate behind a small interface and keep invariants local. Inheritance can be appropriate, but test that derived behavior honors the expectations of callers using the base abstraction.

Check your understanding: Replace the implementation behind an interface and run the same contract tests.

Common trap: Using inheritance only to avoid copying a few lines.

Concept reference: Dev.java: language, collections and JVM learning paths

Not necessarily. Encapsulation protects meaningful invariants and exposes valid operations, rather than merely making fields private. Model state transitions so callers cannot construct invalid combinations, and keep representation details from leaking through mutable references.

Check your understanding: Attempt an invalid transition through the public API.

Common trap: Equating private fields with a complete domain model.

Concept reference: Dev.java: language, collections and JVM learning paths

Look for implementers that must provide meaningless operations or callers that depend on methods they never use. Split responsibilities around real usage while avoiding one interface per line of code. Test consumers against alternate implementations to expose hidden assumptions.

Check your understanding: Implement a second legitimate provider without no-op methods.

Common trap: Designing an interface from one implementation’s internals.

Concept reference: Dev.java: language, collections and JVM learning paths

Lookup depends on the key’s equality and hash behavior. If fields participating in that behavior change after insertion, a later lookup may search a different bucket. Prefer stable value keys, keep equals and hashCode consistent, and test equality with distinct instances representing the same value.

Check your understanding: Insert a key, mutate its equality field, and explain the lookup result.

Common trap: Assuming object identity is always the lookup contract.

Concept reference: Dev.java: language, collections and JVM learning paths

Virtual threads target scalable concurrency for blocking workloads, not extra CPU capacity. Measure the bottleneck and bound scarce resources such as database connections independently. Distinguish task concurrency from throughput and inspect the JDK version and libraries before drawing performance conclusions.

Check your understanding: Compare CPU saturation and database wait time separately.

Common trap: Equating more concurrent tasks with more available CPU cores.

Concept reference: Dev.java: language, collections and JVM learning paths

Enforce the stock invariant at the durable write boundary with a conditional update or appropriate transaction strategy. A process-local lock does not coordinate other instances. Check the affected-row result and treat a losing request as a business conflict rather than reporting success.

Check your understanding: Run two simultaneous purchases and verify stock never becomes negative.

Common trap: Reading the stock and saving a decremented value without an atomic condition.

Concept reference: PostgreSQL: transaction isolation

First, identify technical constraints and define measurable service objectives. Next, follow a request across validation, business logic, storage and asynchronous work. Contrast architectural trade-offs across reuse, clarity, extensibility and accidental complexity, explicitly mitigate the risk of bad data is cached and amplified across downstream consumers, and confirm system stability using semantic validation results, quarantine records and a replay verification.

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 load-test results, retry counts and a measured cost estimate.

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 (follow a request across validation, business logic, storage and asynchronous work), and document the accepted trade-off. Provide concrete proof: a focused unit test, dependency boundary and refactoring comparison.

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 correlate the request trace with transaction state, queue ownership and dependency budgets. Isolate the defect to the smallest reproducible boundary, validate root cause with evidence, and confirm full resolution using segmented service-level indicators, an exemplar trace and an alert threshold.

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

Produce an idempotency test, failure trace and repeatable recovery runbook. Document baseline assumptions, technical mechanism (follow a request across validation, business logic, storage and asynchronous work), rejected alternatives, bounded failure envelopes, and deterministic pass criteria. Supply reproducible verification via queue-depth metrics, bounded retry behaviour and a recovery drill.

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