▸case-01 We are building a financial transaction platform that requires tracking every historical state change without overwriting data, while serving high-throughput read traffic efficiently. Can you design an architectural blueprint for this system, including how commands mutate state, how read views are populated, and how multi-step transfers should be handled across microservices? | fail→pass | 22,705 | 18,698 | -18% | 1 | 1 | 0% | 4,046 | 4,174 | +3% | 0 | 0 | — |
▸case-02 We are creating a basic internal blog where admins post news updates and readers view them. There are no temporal queries, audit trail requirements, or complex workflows. How should we structure the database architecture and service layer? | pass→pass | 13,537 | 13,167 | -3% | 1 | 1 | 0% | 2,568 | 2,908 | +13% | 0 | 0 | — |
▸case-03 We are designing a real-time inventory reservation system for high-frequency trading of concert tickets where two users can never accidentally reserve the same seat simultaneously, requiring strict immediate consistency across all read and write queries without eventual consistency delays. How should we architect this? | pass→pass | 23,557 | 18,233 | -23% | 1 | 1 | 0% | 4,016 | 3,744 | -7% | 0 | 0 | — |
▸case-04 We need a simple health status page that displays whether microservices are currently UP or DOWN. There is no need for historical state reconstruction, compensating actions, or audit trails. How should we store and serve this status data? | pass→pass | 12,565 | 9,382 | -25% | 1 | 1 | 0% | 2,399 | 2,104 | -12% | 0 | 0 | — |
▸case-05 A developer published an incorrect account balance adjustment event into the event stream in production. We need to correct this bad data. Should we run a script to update or delete the corrupted event row in the event store DB, or how should we address it? | pass→pass | 9,796 | 10,190 | +4% | 1 | 1 | 0% | 1,720 | 2,299 | +34% | 0 | 0 | — |
▸case-06 An account aggregate has accumulated over 50,000 historical events over 5 years. Replaying all events from genesis to load the current aggregate state takes several seconds during command validation. How should we optimize command handling latency for this aggregate? | pass→pass | 13,849 | 14,560 | +5% | 1 | 1 | 0% | 2,271 | 3,007 | +32% | 0 | 0 | — |
▸case-07 Our message broker guarantees at-least-once delivery, which means projection workers occasionally receive duplicate event messages. Should we configure the broker for exact-once delivery at the network protocol layer, or how should projection handlers process these events? | pass→pass | 10,353 | 9,833 | -5% | 1 | 1 | 0% | 1,878 | 2,357 | +26% | 0 | 0 | — |
▸case-08 In a microservice system driven by events, debugging production issues across multi-step sagas is difficult because events lack linkage. What metadata fields should be attached to every published event envelope to enable end-to-end transaction tracing? | pass→pass | 16,522 | 13,701 | -17% | 1 | 1 | 0% | 3,106 | 2,988 | -4% | 0 | 0 | — |
▸case-09 We are designing a multi-step order fulfillment workflow across payment, inventory, and shipping services. If the orchestrator server crashes mid-workflow, it loses state and leaves transactions half-completed. What framework architectural pattern should we adopt to persist saga execution state automatically? | pass→pass | 12,590 | 10,165 | -19% | 1 | 1 | 0% | 2,049 | 2,230 | +9% | 0 | 0 | — |
▸case-10 We need to add a new column to a high-traffic read model projection database table in production. Should we alter the existing production read table and backfill it live, or what deployment procedure ensures zero-downtime and safe verification? | fail→fail | 12,749 | 12,985 | +2% | 1 | 1 | 0% | 2,443 | 2,621 | +7% | 0 | 0 | — |
▸case-11 When processing a 'WithdrawMoney' request, should the command handler directly execute an SQL UPDATE on the read-database account balance, or what exact steps must the command handler execute? | pass→pass | 10,892 | 9,558 | -12% | 1 | 1 | 0% | 2,048 | 2,187 | +7% | 0 | 0 | — |
▸case-12 We are defining event names for our domain model. A developer suggests using names like 'UpdateCustomerAddress' and 'ProcessPayment'. Is this appropriate for event sourcing event types, or what naming style should be enforced? | pass→pass | 8,361 | 10,655 | +27% | 1 | 1 | 0% | 1,639 | 2,248 | +37% | 0 | 0 | — |
▸case-13 Our mobile app needs to display a user's dashboard showing order summaries, total spend, and account standing in under 50ms. Querying aggregate event streams directly in response to GET requests is too slow. How should the query layer be structured? | pass→pass | 14,793 | 13,004 | -12% | 1 | 1 | 0% | 2,741 | 2,831 | +3% | 0 | 0 | — |
▸case-14 A compliance regulator asks us to verify the exact state of a user's credit profile as it existed on March 1st at 14:00 UTC. How should our event-sourced architecture handle this temporal query? | pass→pass | 14,024 | 17,783 | +27% | 1 | 1 | 0% | 2,494 | 3,818 | +53% | 0 | 0 | — |
▸case-15 We are changing our 'OrderPlaced' event payload by splitting the 'fullName' string field into 'firstName' and 'lastName' fields. How should we handle existing historical events in the store that use the old single-field schema? | pass→pass | 12,016 | 11,080 | -8% | 1 | 1 | 0% | 2,139 | 2,416 | +13% | 0 | 0 | — |
▸case-16 We are modeling an e-commerce domain with Orders, LineItems, Products, and Customers. Should we place all these entities into a single massive transactional aggregate root to ensure immediate consistency across everything, or how should aggregate boundaries be set? | pass→pass | 11,535 | 14,711 | +28% | 1 | 1 | 0% | 2,213 | 3,145 | +42% | 0 | 0 | — |
▸case-17 During a flight + hotel + rental car booking saga, the rental car reservation fails after the flight and hotel have already been successfully reserved and charged. How should the process manager recover from this failure? | pass→pass | 12,695 | 13,929 | +10% | 1 | 1 | 0% | 2,196 | 3,014 | +37% | 0 | 0 | — |
▸case-18 When selecting or designing our underlying event storage engine, what database access pattern must be strictly enforced for event stream storage? | pass→pass | 7,044 | 6,172 | -12% | 1 | 1 | 0% | 1,147 | 1,455 | +27% | 0 | 0 | — |
▸case-19 How can we ensure that state updates and event publishing to the message broker occur atomically, preventing split-brain scenarios where events are lost if the broker is down when a command finishes? | pass→pass | 13,440 | 12,520 | -7% | 1 | 1 | 0% | 2,549 | 2,640 | +4% | 0 | 0 | — |
▸case-20 A bug in our projection worker calculated customer reward points incorrectly for the past 3 months. The raw event store contains all committed events intact. How do we repair the read model database? | pass→pass | 14,990 | 12,499 | -17% | 1 | 1 | 0% | 2,529 | 2,582 | +2% | 0 | 0 | — |
▸case-21 When implementing snapshotting for performance optimization, when and how often should aggregate snapshots be saved to persistent snapshot storage? | pass→pass | 13,767 | 14,330 | +4% | 1 | 1 | 0% | 2,602 | 2,975 | +14% | 0 | 0 | — |
▸case-22 Two concurrent requests attempt to append new events to the exact same aggregate stream simultaneously. How should the event store detect and handle this concurrent write collision? | pass→pass | 12,057 | 12,026 | -0% | 1 | 1 | 0% | 2,235 | 2,669 | +19% | 0 | 0 | — |