ROLE
Data engineering
Compare expectations across career stages →
These are the same questions found in the library, grouped here by the capability each one mainly examines.
453 questions
Extended questions 251–300 of 453 on this page.
A · Professional problem solving and implementation22
- A study was commissioned to validate a new dashboard. Most users say current reports are sufficient; a few advanced users want more. The project lead wants to highlight supportive interviews. How would you summarize the findings?
- Two requests almost simultaneously check that a business object is absent and then both create it; the application-level check-then-write occasionally produces duplicates. How would you use transaction isolation, unique constraints, and conflict handling to preserve the invariant?
- Given a set of records that may exceed memory capacity, process them in batches while allowing the caller to cancel. Describe or implement an interface that bounds memory, locates errors, and starts no new batch after cancellation.
- Latency is normal after a release, but p99 rises gradually after several hours and temporarily recovers after a restart. CPU is steady while connection counts and heap memory slowly grow. How would you break down hypotheses and design low-risk tests?
- Users can edit a profile on a phone while offline and may also edit it on the web. The device must sync when connectivity returns without silently overwriting valuable changes. How would you design versioning, conflict detection, merging, and user confirmation?
- An entry service calls several downstream systems. One noncritical dependency occasionally slows down, exhausting request threads and affecting unrelated features. How would you design timeouts, concurrency isolation, degradation, and recovery?
- A shared library releases a backward-compatible version, but application repositories cannot upgrade promptly because of pinned dependencies, generated code, or different release windows; a security fix remains blocked. How would you design compatibility policy and the dependency-upgrade process?
- An AI assistant repeatedly calls search, databases, and external APIs for a small share of requests, making latency and cost far higher than average. How would you design per-task budgets, stop conditions, idempotency, and auditable execution records?
- A core service remains within its monthly SLO, but its error budget is being consumed much faster and errors cluster around a recent feature release. How would you decide whether to pause rollout or roll back?
- A new model scores much better on an offline test set, but a small production rollout shows lower task completion for one user group and more complaints. How would you determine where the evaluation set, metrics, or release decision failed?
- An AI assistant retrieves documents from several business systems, where users have different permissions in each. How would you ensure retrieval, caching, summaries, and citations follow source permissions, including after access is revoked?
- The same commit passes reliably on a developer's machine but occasionally fails in CI because of time zones, random ordering, shared caches, or parallel races. How would you classify, reproduce, and reduce this nondeterminism?
- The input contains time intervals that may overlap or touch, and callers need to find records covering a given instant. Describe or implement a merge and query strategy, including whether endpoints are inclusive.
- A streaming service deduplicates events using persistent state and checkpoints. After restoring a checkpoint, the stream may replay events; a crash can also occur between saving the deduplication state and committing the stream offset. How would you make rollback and replay safe while keeping memory bounded?
- The client sees one submission, but the server occasionally creates two identical records. It usually happens after a timeout and retry, with no obvious log error. How would you reconstruct the request timeline and verify where duplication occurs?
- A mobile upload of a large file may be interrupted, and users may retry by tapping again; the system also bills successful processing. How would you design chunks, validation, resumable state, finalization, and the billing point?
- A business-rule correction requires recalculating historical records across several services, but some consumers have already sent notifications or performed external actions. How would you design correction events, impact lists, replay boundaries, and side-effect suppression?
- After a dependency slows down, several callers increase retries at the same time; traffic doubles and the outage spreads. How would you design retry budgets, backoff, jitter, and cross-service coordination?
- A feature is fully launched, but switches created years ago still control several code paths; the team does not know which flags can be removed and relies on memory during rollback. How would you establish flag lifecycles and cleanup ownership?
- A backend API generates TypeScript, Java, and Swift SDKs, but nullable types, enums, and error models are drifting across languages. How would you design the contract source, generation process, and cross-language compatibility tests?
- A production dependency has a different artifact hash from the build associated with its repository tag, and the team cannot confirm it passed the expected CI. How would you establish build traceability, artifact verification, and release authorization?
- An analytics assistant converts natural language into SQL against business data; a request may induce it to read sensitive columns or run an expensive query. How would you design read-only permissions, query validation, resource limits, and result redaction?
No questions in this section match your target. ·
C · Value judgment and direction setting19
- How is your core metric defined? What are the numerator, denominator, time window, de-duplication, and exclusion rules, and why is it the core metric?
- How was your funnel built? Why are the steps divided the way they are, where does most of the drop-off happen, and how do you confirm the steps neither overlap nor leave gaps?
- How do changes at each funnel stage ultimately act on the business metric? Improving which stage by a little moves the business metric the most, and what is that judgment based on?
- Once your metric is converted to the industry's common definition, where does it stand? What assumptions did the conversion make, and what information did it lose?
- A dashboard holds dozens of metrics and the team's attention keeps getting pulled to the most visible numbers. Pick one user task and explain how you would tier the metrics and decide which enter the core list.
- The metric was met, but you felt the business did not really get better. Has that happened to you? How did you notice, and what did you do?
- After a feature launches, how do you judge whether it met its original goal? When do you review it, and what would you decide differently afterwards?
- For the last new feature you launched, what were the specific before-and-after changes in the core metric, such as conversion, click-through, or retention? If the numbers fall short of expectation, along which dimensions do you usually investigate the cause?
- A feature encourages invitations, so inviters and friends may land in different experiment groups. Results are significant, but spillover cannot be fully isolated. How would you assess the conclusion?
- User mix or external conditions change during a project, making the original baseline incomparable. How would you continue judging the outcome?
- The target metric improved, but refunds, complaints, or manual handling also worsened. How would you decide whether the outcome is acceptable?
- The initial effect of a feature is strong but fades over several months. How would you distinguish novelty from lasting behavior change?
- The final business outcome will take months to appear, but the team must decide whether to continue investing now. How would you use leading indicators?
- Several teams changed the product, pricing, and operations at once, and the overall outcome improved. How would you assess each contribution?
- An experiment shows no expected lift, but also does not prove the idea ineffective. How would you decide whether to continue, change, or stop?
- Two projects each meet some goals, but one delivers more user benefit per unit of investment. How would you reallocate resources?
- A pilot meets its target, but results vary when expanded to more regions or segments. How would you judge whether the result is replicable?
- A new process doubles weekly feature releases, but task completion is unchanged and support tickets rise slightly. How would you decide whether to continue, adjust, or stop?
- After a redesign of an important flow, the expected completion rate does not improve. Support requests decline, existing users respond more positively, but new users behave worse. How would you check metric definitions and user segments, determine what effect the design had, and choose the next action?
No questions in this section match your target. ·
D · Delivery and ongoing operations9
- Different teams provide conflicting data and explanations for the same business problem, while the project is already planned for this quarter. How would you decide what to investigate first and what to defer?
- A campaign goes live overnight, but heavy traffic stalls the system. Users cannot claim coupons and start complaining. As an engineer involved in the campaign, how would you coordinate with operations and support on temporary business measures while restoring the system?
- Describe a re-engagement or referral campaign you contributed to as an engineer. Which systems, data, or delivery work did you own, how did you clarify budget and key user behavior with operations, and how did you verify incremental impact and your actual contribution?
- The promotion strategy is set and you are responsible for execution. What goals, rules, resources, risks, and acceptance criteria would you confirm before launch?
- When managing a large platform ecosystem, gray areas such as fake orders, abusive bonus hunting, and borderline content often appear. How do you build a governance mechanism that does not harm the healthy ecosystem, and how do you build the defense together with risk control, legal, and technology teams?
- You are about to run a growth experiment in production. What checkpoints and stop conditions do you prepare first?
- How would you move from one-off campaigns and acquisition toward automated, data-informed lifecycle operations? Describe a case where data tracking and algorithms turned manual work into repeatable rules.
- A two-sided platform has a local supply-demand imbalance: peak-time orders exceed available providers, or a new city has providers but too few users. How would you identify the main constraint and adjust pricing, incentives, or supply while controlling cost and long-term effects?
- A strategic change requires you to shut down at once a business that brings many users but has lost money for years. How do you, in the shortest time, reassure thousands of affected partners, complete the wind-down compliantly and smoothly, and prevent a group public-relations crisis?
No questions in this section match your target. ·