Puzzle questions have largely fallen out of favour in technical hiring, and for good reason — they measure whether you have seen the puzzle before more than how you think.
They remain interesting as puzzles, though, and the reasoning patterns are genuinely portable: bound the problem, look for invariants, ask what information each step actually buys you. Treat these as recreation rather than interview preparation.
Every question below shows all four options, the keyed answer, and the one-line explanation. If you want to turn these into video, the OmniFrame Quiz Studio is free and this pack is built in.
Question 1
Which workload detail helps choose between fan-out-on-write and fan-out-on-read for a large social feed?
Why: The balance of reads, writes, and follower distribution drives the appropriate feed fan-out strategy.
Question 2
A cache improves latency but causes wrong results after updates. What policy is missing?
Why: Cached data must be expired or refreshed when source data changes.
Question 3
A search feature is slow because it scans every row. What structure helps?
Why: Indexes speed lookup by organizing searchable fields.
Question 4
A service works in one region but fails globally. What did the design ignore?
Why: Global systems need redundancy and failover.
Question 5
Your system accepts payments twice during retries. What guarantee do you need?
Why: Idempotent operations prevent duplicate side effects.
Question 6
How would you estimate the number of piano tuners in a city?
Why: Break the problem into population, households, piano ownership, and service frequency.
Question 7
What data structure gives fast lookup by key?
Why: Hash tables map keys to values with average constant-time access.
Question 8
What is the tradeoff of caching data?
Why: Cached data is faster but can become stale.
Question 9
Why use a queue in a system design problem?
Why: Queues smooth spikes and decouple producers from consumers.
Question 10
What is binary search's key requirement?
Why: Binary search repeatedly halves a sorted search space.
Question 11
What does Big O describe?
Why: Big O describes how time or memory grows with input size.
Question 12
What is the fastest way to detect duplicates in a list?
Why: A set tracks seen values efficiently.
Question 13
Why shard a database?
Why: Sharding distributes data across machines.
Question 14
What failure should every distributed system expect?
Why: Networks can be slow, partitioned, or unavailable.
Question 15
What is the product manager version of an algorithm tradeoff?
Why: Engineering choices should serve user outcomes, not just elegance.
These questions are free to use — in a classroom, a pub quiz, a video, or anywhere else. No attribution required. If you think one is ambiguous or an explanation is wrong, tell us and we will fix the page.
For how questions like these are constructed, see what makes a quiz question hard.