ROLE
AI engineering
Compare expectations across career stages →
These are the same questions found in the library, grouped here by the capability each one mainly examines.
1175 questions
Extended questions 951–1000 of 1175 on this page.
A · Professional problem solving and implementation11
- Model responses may echo customer identifiers, while production debugging relies on request and response logs. How would you design collection, redaction, access, and retention?
- Users see response times double, but model inference time has barely changed. How would you break down the request-processing stages, locate the bottleneck, and limit impact?
- The team tunes prompts and model settings against the same fixed evaluation set every week. Scores rise, but new-user feedback does not improve. How would you restore evaluation trustworthiness?
- A model passes evaluations in a major market but behaves inconsistently in a new region because language, terminology, and service rules differ. How would you set rollout scope and follow-up gates?
- To reduce inference cost, the team caches answers by prompt text. User permissions, knowledge versions, and regional rules may differ. How would you define cache keys and invalidation?
- The evaluation set has grown, but many items are slight rewrites of one template, making model scores look stable. How would you audit sample quality?
- A natural-language request omits an amount limit and target account, while the agent can invoke a side-effecting tool. How would you decide whether to clarify, advise only, or require confirmation?
- A shared cache is read-heavy and the team plans to replace a mutex with a reader-writer lock. What evidence would you gather, and how would you judge whether the added complexity is worthwhile?
- A migration across millions of records cannot fit in one transaction. The business cannot expose a partially completed state. How would you design batched writes, visibility, and recovery?
- A service accumulates monetary values with floating-point numbers, and reconciliation drift grows over time. How would you identify the source and redesign calculations while migrating historical data?
- A feature targets people who have never used the product, but all usability participants are core users. Which design decisions can the current evidence support?
No questions in this section match your target. ·
B · Business understanding and domain modeling5
- A service rebuilds state from an event log, but the live snapshot differs and responders cannot tell which to repair. How would you define authority and the repair process?
- A promotion starts at local midnight, services run in UTC, and customers may move regions. How would you represent effective time and handle boundary invoices?
- After a feature definition is fixed, the modeling team wants a historical backfill. Training reproducibility requires the original version, while business reports want corrected values. How would you design both views?
- Risk teams need to reproduce decision inputs from three months ago, but the online feature store keeps only the latest state to save cost. How would you balance reproducibility and storage cost?
- High-revenue customers frequently request urgent support, consuming team capacity and delaying critical issues for others. How would you define service prioritization?
No questions in this section match your target. ·
C · Value judgment and direction setting4
- Task completion reaches 95%, but half of users rely on support or colleagues. The team plans to use completion as its launch-success metric. What would you add?
- The same customer submits changing feature requests, all pointing to delayed month-end reconciliation. How would you uncover the underlying problem and validate priority?
- Three teams propose shared-capability projects, each citing different future benefits with no near-term customer commitment. How would you validate and sequence platform investments?
- The team completed its top-ranked projects, but engineers with critical skills are idle while other teams are overloaded. How would you improve allocation?
No questions in this section match your target. ·
D · Delivery and ongoing operations13
- A project is nearly complete, but the business sponsor has left and the receiving team has no clear owner. How would you handle acceptance, handoff, and unresolved risk?
- The team has been interrupted by unplanned support work for three quarters and commitment completion has fallen. Leadership wants everyone fully booked to raise utilization. How would you propose a more reliable plan?
- An outage did not meet the threshold for a major customer incident, but affected a small group and exposed process gaps. How would you decide whether and how much to publish?
- Three projects need an integration environment that cannot be shared concurrently. Release dates are close and they depend on the same test team. How would you set scheduling rules and contingencies?
- The team believes rework is excessive, but some count bug fixes, others requirement changes, and others code rewrites. How would you define the problem and set an improvement target?
- The same remediation command has different effects in two regions, but the runbook describes only the default topology. Responders may use it in the wrong environment. How would you revise and prevent recurrence?
- Upstream events are delayed by hours, while downstream reports still show success and customers make incorrect decisions. How would you design freshness states and notifications?
- A vendor’s committed throughput is below peak-season forecast, and migration to an alternative takes two months. How would you plan capacity and business constraints?
- Disaster-recovery drills notify every team a week in advance, and special scripts make the exercise succeed while real recovery still fails. How would you redesign the drill?
- Scope is frozen and acceptance is underway when compliance identifies a new mandatory requirement. Releasing as planned creates audit risk. How would you respond?
- The monthly cloud bill is under budget, but cost per transaction keeps rising; temporarily lower volume hides efficiency regression. How would you govern cost?
- The business depends on a third-party API with no formal SLA and inconsistent support. How would you set internal continuity goals and fallback options?
- Leadership asks for higher engineering productivity next quarter without defining it. Some propose increasing commits; others want fewer meetings. How would you establish an improvement direction?
No questions in this section match your target. ·
E · Collaboration, communication, and influence2
- Project documents are fully accessible, but meetings repeatedly revisit decided options and newcomers do not know what can be reopened. How would you improve decision records?
- A dependency team verbally promises a quarterly interface but will not assign an owner, capacity, or milestones. Your project has already committed externally. How would you renegotiate?
No questions in this section match your target. ·
F · People and organization building7
- A normally reliable teammate has slowed down while carrying three cross-team projects and mentoring a newcomer. How would you diagnose and respond?
- The organization expects a fixed number of promotions this quarter, but candidates show uneven scope and impact evidence. How would you facilitate calibration?
- After a key leader leaves, a potential successor handles most responsibilities, but pay, decision rights, and external title have not changed. How would you manage the transition?
- A critical vacancy delays delivery and the business wants an offer within a week. A candidate has strong experience but unclear integrity signals. How would you balance speed and standards?
- Interviewers give referred candidates more benefit of the doubt for vague answers and score others strictly. How would you reduce this bias?
- A first-round interviewer writes “not capable,” so later interviewers only probe weaknesses. How would you design evaluation and information sharing?
- After a reorganization, projects continue but interfaces and informal coordination links disappear, and issues bounce between teams. How would you restore accountability?
No questions in this section match your target. ·
G · Learning, adaptation, and sustained growth5
- A teammate repeatedly delivers through crises, but recovery takes longer each time and routine errors are increasing. How would you respond?
- The team correctly identifies an incident’s root cause and records improvements, but operating habits have not changed months later. How would you evaluate the retrospective?
- The organization has no clear promotion timeline, but you want to keep growing. How would you choose learning, influence, and delivery goals without chasing visibility alone?
- Several teams have adopted a plan you championed, then you discover a key assumption is false and continuing will increase rework. How would you correct course?
- A project was cancelled for insufficient market demand, but its technical capability may be reusable. How would you decide what to retain, transfer, or stop investing in?
No questions in this section match your target. ·
H · Professional responsibility and work style3
- A quarterly update describes correlation as causal growth. The owner wants to publish on time and validate later. How would you respond?
- A role offers significant growth but repeatedly requires night responses and shifting priorities. How would you assess staying, renegotiating, or leaving?
- After proactively taking on a cross-team issue, your original project begins slipping. The team also depends on you for the new issue. How would you address priorities and boundaries openly?
No questions in this section match your target. ·