React Developer DBMS Concepts Interview Guide

At Read Committed, each statement receives a new snapshot, so later statements can see commits made in between. Higher isolation changes the guarantees and may require retrying serialization failures. Choose isolation from the business invariant, not from an assumption that every transaction freezes all reads.

Check your understanding: Interleave two sessions and document the snapshots each statement sees.

Common trap: Treating transaction atomicity and isolation as identical.

Concept reference: PostgreSQL: transaction isolation

An index is an option, not a command. The planner considers selectivity, statistics, table size and access cost. Read estimated and actual rows, filters and scan types before changing indexes. EXPLAIN ANALYZE executes the statement, so use safe data and transaction handling for writes.

Check your understanding: Compare estimated and actual row counts on representative data.

Common trap: Forcing an index without measuring the full query.

Concept reference: PostgreSQL: reading EXPLAIN plans

Check the grain of each table and whether the join creates several rows for one order. Aggregate at the intended business grain before joining, or use the correct relationship and key. Reconcile totals using a tiny fixture containing multiple child rows and an unmatched parent.

Check your understanding: Use one order with two child records and verify the total remains correct.

Common trap: Using DISTINCT as a universal repair for a wrong join.

Concept reference: PostgreSQL: reading EXPLAIN plans

If filtering depends only on current items and the search term, calculate it during rendering. A second state variable and an Effect introduce a render with stale data and another update. Measure expensive filtering before adding memoization. Keep Effects for synchronization with external systems, with cleanup for subscriptions.

Check your understanding: Change the search term and data independently; verify there is no stale intermediate list.

Common trap: Treating useEffect as the default place for every calculation.

Concept reference: React: when an Effect is unnecessary

Choose one owner at their nearest shared boundary and pass the value and change handlers down. Keep transient UI state local where possible. For remote persistence, distinguish the editable draft from the last saved version so a late response cannot overwrite new typing.

Check your understanding: Edit from both components and simulate an out-of-order save response.

Common trap: Duplicating an editable value in several components without a reconciliation rule.

Concept reference: React: when an Effect is unnecessary

Exercise the form through its labels and submit control. Simulate a failed request, assert that the draft and accessible error remain, then retry and verify the success state. Separate server validation from client hints and test repeated clicks rather than checking component internals.

Check your understanding: Run the same scenario using keyboard navigation and a delayed response.

Common trap: Only asserting that a mock function was called.

Concept reference: Playwright: reliable browser tests

First, identify technical constraints and define measurable service objectives. Next, trace state, rendering and network work from user action to painted result. Contrast architectural trade-offs across write cost, read latency, consistency and operational 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 lease-expiry tests, idempotency records and a restart recovery drill.

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 state, rendering and network work from user action to painted result), and document the accepted trade-off. Provide concrete proof: an explain plan, concurrency test and recovery verification.

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 inspect the interaction trace, render causes, network waterfall and accessibility tree. Isolate the defect to the smallest reproducible boundary, validate root cause with evidence, and confirm full resolution using a compatibility test matrix and a staged rollout metric.

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

Produce a performance profile and keyboard-accessible regression test. Document baseline assumptions, technical mechanism (trace state, rendering and network work from user action to painted result), 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.