Aurora coding interview
questions, leaked.
4 problems reported across recent Aurora interviews. Top patterns: hash table, depth first search, linked list. The list below is what most reported candidates actually saw, plus the honest play if you can't grind all of it.
Aurora's assessment skews toward hash tables and graph traversal. You're looking at 4 problems across Easy to Hard, with LRU Cache and Number of Distinct Islands dominating the reported list. Both require solid fundamentals in data-structure design and DFS/BFS. The good news: these patterns repeat, so drilling them now pays off. The better news: if you hit a wall during the live assessment and blank on how to structure an LRU Cache or optimize island counting, StealthCoder runs invisibly and surfaces a working solution in seconds.
Top problems at Aurora
| # | Problem | Diff | Frequency | Pass % | Patterns |
|---|---|---|---|---|---|
| 01 | LRU Cache | MEDIUM | 100.0 | 45% | Hash Table · Linked List · Design |
| 02 | Number of Distinct Islands | MEDIUM | 74.6 | 62% | Hash Table · Depth-First Search · Breadth-First Search |
| 03 | Diameter of Binary Tree | EASY | 74.6 | 64% | Tree · Depth-First Search · Binary Tree |
| 04 | Word Search II | HARD | 74.6 | 37% | Array · String · Backtracking |
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 Aurora OA. The proctor cannot see it. Screen share cannot detect it. Made by a working Amazon engineer who got tired of watching qualified friends bomb OAs they'd solve cold in an IDE.
Get StealthCoder- hash table2 · 50%
- depth first search2 · 50%
- linked list1 · 25%
- design1 · 25%
- doubly linked list1 · 25%
- breadth first search1 · 25%
- union find1 · 25%
- hash function1 · 25%
- tree1 · 25%
- binary tree1 · 25%
Hash tables appear in half the reported problems. That's not coincidence. Depth-first search shows up twice as well, often paired with hash logic or graph problems. LRU Cache demands you combine hash tables with doubly-linked lists, which is a real implementation test, not a theory question. Number of Distinct Islands requires you to choose between DFS, BFS, or union-find, then justify the choice. Diameter of Binary Tree is the gimme: standard DFS recursion. Word Search II is the trap: it needs trie construction and backtracking, a multi-step problem that eats time. Go in confident on tree traversal. Be ready on hash-table design patterns. And if you get stuck mid-OA on the island-counting variation or cache eviction logic, StealthCoder is your silent backup to unblock you.
Companies with similar patterns
If you prepped for Aurora, these companies recycle ~60% of the same topics.
You've seen the list.
Now make sure you pass Aurora.
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. Made by a working Amazon engineer who got tired of watching qualified friends bomb OAs they'd solve cold in an IDE. Works on HackerRank, CodeSignal, CoderPad, and Karat.
Aurora interview FAQ
How much time should I spend on hash-table problems vs. graph traversal?+
Split it 50-50. Hash tables appear in 2 of 4 reported problems, often combined with other patterns. DFS appears twice. Master LRU Cache first because it's the most complex design, then drill DFS patterns on tree and island problems. You can't skip either.
Is Diameter of Binary Tree enough DFS practice for Aurora?+
No. It's the easy one. Number of Distinct Islands requires DFS or BFS with spatial reasoning. That's harder. Drill both the tree recursion and the 2D grid traversal. Don't assume the medium and hard problems use the same DFS logic.
Should I worry about Word Search II if I only have a week?+
Yes. It's marked Hard for a reason: trie construction plus backtracking. You can't code it from scratch in 10 minutes. Study trie insertion and DFS backtracking separately, then combine them. If time is tight, this is the one where StealthCoder saves you during the actual assessment.
Do I need union-find for Number of Distinct Islands, or is DFS sufficient?+
DFS is sufficient and faster to code under pressure. Union-find is an alternative mentioned in the reported problems, but it's overkill for island counting. Know DFS cold. Understand union-find as a backup pattern, not a must-drill.
What's the order to study Aurora's problems?+
Start with Diameter of Binary Tree (easy, builds DFS confidence). Then LRU Cache (design + hash tables, medium). Then Number of Distinct Islands (DFS on grids). Save Word Search II last. That order builds momentum and skills progressively.