Loading
Pick a focused question that fits your time, stack, and interview goal.
690 results
Page 3 of 29
Uses an ordered window aggregate to calculate a deterministic running balance per account.
Implements invoice total calculation with quantity, unit price, percentage discount, and rounding.
Combines daily aggregation, a calendar table, and a window frame to calculate true seven-calendar-day revenue.
Design background jobs that handle cancellation, retries, and duplicate execution without corrupting external state.
Use CancellationToken to stop wasted work when clients disconnect, timeouts expire, or background operations are being shut down.
Explain why CancellationTokenSource lifetime matters and how careless timeout creation can leak timers or resources.
Explains dependency duration, error rate, timeout rate, query tagging, Redis latency, HTTP downstream calls, and endpoint correlation.
Use Channel<T> to coordinate producer/consumer background processing inside a .NET service while respecting backpressure and shutdown.
Chooses contracts and shared implementation based on substitutability, state ownership, evolution, and inheritance constraints.
Explains partition key selection for ordering, load distribution, hot keys, and consumer scalability.
Builds a story about resisting unnecessary complexity and choosing maintainable design.
Explains how TTL choices balance freshness, load reduction, user expectations, and operational safety.
Chooses heap-focused GC evidence or complete process state based on memory growth, roots, hangs, crashes, locks, and collection risk.
Chooses a sequential collection from size, mutation, access pattern, memory locality, and the behavior callers actually need.
Avoids hidden quadratic work by indexing repeated matches while preserving duplicate, missing, ordering, and memory semantics.
Compares test doubles by purpose and explains when a real dependency provides a more trustworthy test.
Chooses a type shape from identity, value semantics, mutation, copy cost, default values, and serialization boundaries.
Maps business invariants and stale-data tolerance to strong or eventual consistency, explicit failure behavior, reconciliation, and user-visible state.
Creates valid objects with clear intent while choosing construction patterns proportionally to optional data, validation, and lifecycle complexity.
Reason through mitigation choices during a live App Service production incident.
Maps creation, updates, asynchronous work, validation, authorization, conflicts, and failures to useful HTTP outcomes.
Chooses constraints, conditional writes, optimistic conflicts, or stronger isolation to protect a limited-resource reservation without unnecessary blocking.
Explains when optimistic UI is worth it, when to wait for server confirmation, and how to recover from rejected mutations.
Places state according to who owns it, how long it must live, and whether the URL or server is the source of truth.