Affirm coding interview
questions, leaked.
11 problems reported across recent Affirm interviews. Top patterns: array, hash table, string. The list below is what most reported candidates actually saw, plus the honest play if you can't grind all of it.
Affirm's interview filters hard on arrays, hash tables, and strings. You'll see 6 problems per topic across 11 total questions, split 2 easy, 7 medium, 2 hard. That's a stacked middle tier. Design problems show up frequently too, which means you're not just grinding brute force. If you haven't built a data structure from scratch under time pressure, that's the gap. StealthCoder runs invisibly during your live assessment, so if you hit a wall on Insert Delete GetRandom or Parse Lisp Expression mid-OA, you get a working solution in seconds.
Top problems at Affirm
| # | Problem | Diff | Frequency | Pass % | Patterns |
|---|---|---|---|---|---|
| 01 | Insert Delete GetRandom O(1) | MEDIUM | 100.0 | 55% | Array · Hash Table · Math |
| 02 | Shortest Uncommon Substring in an Array | MEDIUM | 81.0 | 48% | Array · Hash Table · String |
| 03 | Group Anagrams | MEDIUM | 76.6 | 71% | Array · Hash Table · String |
| 04 | Parse Lisp Expression | HARD | 74.9 | 53% | Hash Table · String · Stack |
| 05 | Design Hit Counter | MEDIUM | 73.0 | 69% | Array · Binary Search · Design |
| 06 | Insert Delete GetRandom O(1) - Duplicates allowed | HARD | 70.9 | 36% | Array · Hash Table · Math |
| 07 | String Compression | MEDIUM | 68.4 | 58% | Two Pointers · String |
| 08 | Valid Anagram | EASY | 65.6 | 67% | Hash Table · String · Sorting |
| 09 | Coin Change | MEDIUM | 58.3 | 46% | Array · Dynamic Programming · Breadth-First Search |
| 10 | String Compression III | MEDIUM | 45.8 | 67% | String |
| 11 | Number of Recent Calls | EASY | 45.8 | 77% | Design · Queue · Data Stream |
Frequencies derived from public community-tagged interview reports. Click a row to view on LeetCode.
You have a week, maybe less. You can't out-grind the list above. StealthCoder runs invisibly during the actual Affirm OA. The proctor cannot see it. Screen share cannot detect it. Built by an Amazon engineer who realized the OA tests how well you memorized 200 problems, not how well you code.
Get StealthCoder- array6 · 55%
- hash table6 · 55%
- string6 · 55%
- design4 · 36%
- math2 · 18%
- randomized2 · 18%
- queue2 · 18%
- data stream2 · 18%
- sorting2 · 18%
- binary search1 · 9%
Array and hash-table problems dominate equally, and they're often combined. Shortest Uncommon Substring, Group Anagrams, and Valid Anagram all test your fluency with both simultaneously. String shows the same frequency, which means you can't skip compression or character-frequency drills. The two hard problems are design-heavy (Insert Delete GetRandom with duplicates, Parse Lisp Expression), so one of your two hardest questions will likely require you to think architecturally, not just code a loop. Design Hit Counter and Number of Recent Calls teach you to think in queues and time windows. Start with hash-table plus array combos, then nail string problems, then move to design. StealthCoder is your safety net if a design constraint blindsides you during the actual OA.
Companies with similar patterns
If you prepped for Affirm, these companies recycle ~60% of the same topics.
You've seen the list.
Now make sure you pass Affirm.
Memorizing every problem above in a week is a fantasy. StealthCoder is the hedge: an AI overlay that's invisible during screen share. It reads the problem on screen and surfaces a working solution in under 2 seconds. Built by an Amazon engineer who realized the OA tests how well you memorized 200 problems, not how well you code. Works on HackerRank, CodeSignal, CoderPad, and Karat.
Affirm interview FAQ
Should I spend more time on arrays or hash tables for Affirm?+
Both equally. They appear 6 times each in their pool and are almost always paired. Hash tables let you solve array problems faster, so start hash table first, then practice combining them. Group Anagrams and Shortest Uncommon Substring are your templates.
How many design problems should I prepare for?+
Four design problems appear in Affirm's data. Insert Delete GetRandom (both versions) and Design Hit Counter are must-knows. Practice building simple data structures with O(1) operations, because at least one question will demand architectural thinking, not just iteration.
Is string really that important for Affirm?+
Yes. Six string problems in 11 total questions. Valid Anagram, String Compression, Parse Lisp Expression, and Shortest Uncommon Substring cover the range. If you're weak on character counting or recursion-based parsing, drill those now before your OA.
What's the difficulty split I should expect?+
Mostly medium (7 of 11). Two easy warm-ups, two hard problems that will likely be design-based. The mediums are where you prove you can code fast and clean. Don't overweight hard-problem prep; nail the mediums under time pressure.
Do I need to know binary search and tries for Affirm?+
Binary search appears once (Design Hit Counter), trie once (Shortest Uncommon Substring). Not core focus, but Design Hit Counter combines binary search with queue logic, so review it. Tries are nice-to-have but lower priority than arrays and hash tables.