Project Delivery and Accountability Questions

How you ensure projects stay on track, handle scope creep, communicate delays to leadership, recover from setbacks, balance technical excellence with delivery, and take ownership of outcomes even when multiple factors are involved.

MediumTechnical

0 practiced

Write a PostgreSQL SQL query to compute weekly active users (WAU) per ISO week for the last 12 weeks given this events table.

events(
  event_id SERIAL PRIMARY KEY,
  user_id INT,
  event_type TEXT,
  event_time TIMESTAMP WITHOUT TIME ZONE,    -- when the event occurred
  received_time TIMESTAMP WITHOUT TIME ZONE  -- when the event was ingested into the warehouse
)

Requirements: return week_start_date, wau. Account for late-arriving events that can be received up to 48 hours after event_time (include events where received_time <= week_end + interval '48 hours'). Only count unique user_id per week. Provide a query robust to duplicates and explain assumptions.

HardSystem Design

0 practiced

Design a policy for metric definition and change control (a metric catalog). Define who can propose changes, how to test changes (automated tests, shadow runs), versioning, how changes are published to dashboards, rollback criteria, and the human approvals required. Include automated tests you would run before publishing.

EasyTechnical

0 practiced

List and briefly describe the typical BI project lifecycle phases from request intake to production dashboard maintenance. For each phase, name one artifact you produce (example: intake form, requirements doc, data lineage, acceptance criteria, runbook) and one common risk to watch for during that phase.

EasyTechnical

0 practiced

You're late delivering a high-visibility dashboard by two days. Draft the communication you'd send to leadership: explain the delay succinctly, root cause, immediate mitigation, expected new delivery date, and the actions you'll take after delivery to restore confidence and prevent recurrence.

HardTechnical

0 practiced

Technical: You need to incrementally rebuild a denormalized reporting table nightly, but source systems don't provide CDC flags and full history reprocessing is expensive. Describe an approach (SQL or pseudocode) for detecting changed rows, handling deletions, and keeping the reporting table accurate. Include performance trade-offs and how you'd schedule the rebuild to minimize cost.

Unlock Full Question Bank

Get access to hundreds of Project Delivery and Accountability interview questions and detailed answers.

Join thousands of developers preparing for their dream job.