Amazon NG VO contains 3 rounds back-to-back, each round lasting 1 hour. There was a 30 minute break between the second and third rounds.
First round of BQ + coding, the second round of pure coding, the third round of pure BQ.
Amazon VO BQ Question：
1. Tell me about a time you failed / The biggest mistake you made. （Ownership）
2. Tell me about a time when you had to work on a project with unclear responsibilities.（Bias for Action）
3. Tell me about a time when you gave a simple solution to a complex problem.（Invent and Deep Dive）
4. Understand complex problems and solve them with a simple solution.
5. Make an easier implementation for customers.
Amazon VO Coding
Q1: LRU Cache. The key is implementing a LinkedHashMap. The underlying data structure consists of a hash table plus a doubly linked list.
Q2: Design a custom data structure to evaluate expressions of the form (2*(3+4*(1+2))). You need to account for a memory allocator to ensure safety, map operation tokens to data-structure methods, and manage object lifecycles by providing a free method to avoid memory leaks.
Q3: A simple hash‐table and frequency‐matching problem—a basic LeetCode question. My first thought was a brute‐force approach, but that leads to O(n2)O(n^2)O(n2) time complexity, which is too slow for large arrays and likely to be challenged by an interviewer. Building a hash table to track frequencies optimizes the solution.
Q4: “Clone Graph,” a standard LeetCode problem. You need to handle each node’s value and its neighbor relationships while avoiding revisiting the same node. Use a hash table to map original nodes to their clones, preventing duplicates. You can traverse the graph with either DFS or BFS.
For More Information
Through our strong Amazon NG VO assisted interview , candidates not only demonstrate coding skill but also clear thinking and effective communication. Good luck to everyone!
