BigOBigOCodes LogoCodes
  • Articles
  • VO面试
  • 面经
  • NIW论文
Back to articles
October 1, 2025|4 min read

亚麻6月VO面经分享(含行为问题 + 面向对象设计题)

以下为本人真实的面试经验,用了一份NG的简历套的面试,亚麻的技术面通常伴随BQ,这一轮是BQ+OOD,整体流程大约持续 1 小时面的AWS某个百人组。这一轮是一个烙印。


Learn more about interview preparation

想知道更多的
BigOCodes面试辅助?

了解更多联系我们

Keep Reading

View All
meta e4 interview prepare
meta e4 interview prepare

Meta E4 面试你将会面对什么?

买它到底面啥?流程几轮?哪些地方容易挂?需要准备什么?要写代码?要系统设计?要讲故事?HR 到底在意啥?奋笔血书 6 小时写出来的全流程经验总结,从 Intro Call 一直到 VO 的每一轮 Coding、System Design、BQ 行为问题,全都拆开说。

May 25, 2025Read
meta may summary
meta may summary

Meta买它5月算法高频题分析

一直在整理地里的分享出来的Meta 面经,最近一个月的算法考点是什么?你又碰到原题吗?5月算法收录了近 70 道,最多的一题346出现了5次。meta应该是快结束招人了,相比3月数量小了很多。

June 4, 2025Read
what is codex interview
what is codex interview

大O代码自研面试利器对Zoom共享屏幕完美隐身

AI面试工具层出不穷,但AI真的懂面试吗?发展到现在为什么Bigocodes还在研发依赖人工的面试工具?一篇技术文章告诉你什么是Codex Interview?我们为什么要开发Codex Interview?

July 21, 2025Read

Advisor

Jim

Meta 软件工程师, 设计和开发基于服务器端技术 Phd candidate of UC Berkely

Yuxuan

AMD 全栈工程师, 开发和维护AMD 产品的前端和后端系统 M.S of UTexas, Austin

Liu

Tesla 数据分析师, 收集分析和解释车辆及驾驶数据 M.S of University of Southern California

Collaborator

Benjamin
Olivia --B.S of Syracuse University
Wu --B.S of UC Berkely

官方微信

Taobao官方淘宝店铺Kris努力搞副业
©BigOCodes 2026

第一部分:BQ

Q1: Tell me about a time you took on something outside your area of responsibility. (Positive问题)

这是一个关于主动性与“主人翁精神”(Ownership)的典型问题。你可以讲述一个你明明不需要做、但你发现问题并主动解决的故事。

Q2: How do you approach learning a new technology or unfamiliar domain? (Positive问题)

考察学习能力与“乐于学习与好奇心”(Learn and Be Curious)。可以分享你如何自学新框架、新工具,并快速应用到项目中。

Q3: Tell me about a time where no existing solution worked and you had to create your own. (Positive问题)

强调你的创新能力与“发明并简化”(Invent and Simplify)的精神。比如你如何规避现有限制、提出新思路并成功落地。

Q4: Describe a time you received constructive/critical feedback and how you handled it. (Negative问题)

考察“赢得信任”(Earn Trust)与抗压成长的能力。讲一个你起初不被认同,但最终通过沟通、反馈、改进赢得团队信任的例子。

Q5: Tell me about a time when you thought you wouldn’t meet a commitment.(Negative问题)

这是一个关于“结果导向”(Deliver Results)的压力测试题。你可以分享一个计划被打乱时如何调整资源或目标,最终还是交付的故事。


第二部分:OOD

这一部分亚麻给的题都很和实际相关,我其实挺喜欢这样的题的。太抽象了不好理解。

面试题原文如下:

For most packages, the primary tasks we need to perform are picking up the package from some location and transporting it to its drop-off location. However, it’s possible that picking up the package fails—perhaps the driver couldn’t find the package. In that case, the package can no longer be dropped off. Let’s capture this in our Task and Route model.

Please:

  1. Update your Task model to capture “Pickup” and “Drop-off” tasks for a package, and the relationship between them.
  2. Update your Route to support marking the current task as failed/success.
  3. Update your Route to skip over Tasks that cannot be executed during “getCurrentTask”, based on failures of their upstream tasks.
    • E.g. a failed pickup task should result in skipping over the drop-off task for the same package.

Assume that the list of tasks provided to you has all of the tasks that might be executed in the order in which they appear — if they aren’t cancelled.

Example Input Task List:

(pickup package a)
(pickup package b)
(drop-off package a)
(drop-off package b)

Simulation Output:

getCurrent -> pickup package a  
mark failure  
getCurrent -> pickup package b  
mark success  
getCurrent -> drop-off package b  
// note: drop-off package a is skipped

解题思路:

  • 将 Task分为两类:PickupTask 和 DropoffTask,它们都与同一个 Package 实例相关联。
  • 每个任务需要有 status状态字段,例如:PENDING、SUCCESS、FAILURE。
  • Route 类负责维护一个任务列表 tasks[],并追踪当前进行到哪个任务(可以用索引或队列)。
  • 在 getCurrentTask() 中,如果任务是 drop-off 类型,但对应的 pickup 任务失败,就要自动跳过。
  • 每次调用 completeCurrentTask(status) 后,推进到下一个可执行任务。

这题虽然不涉及复杂算法,但核心在于:

  • 设计出 合理的类结构与职责分工;
  • 处理任务间的 状态依赖关系;
  • 能将现实逻辑抽象成清晰的代码流程。

如果你正在准备 Amazon 的技术面试,这一轮非常贴近实际业务开发,强烈建议练习类似问题并准备好 STAR 故事来对应 Amazon 的 Leadership Principles。