提问 HN：操作系统内核团队的入职培训是什么样的？
7 分•作者: markus_zhang•7 个月前
背景：我想知道操作系统内核团队是如何对其新员工进行培训的。
我主要感兴趣的是：
* NT 内核 / XNU 内核 / Android 内核团队，或者任何非开源的，比如一些专有的 RTOS 团队也可以算。
* 当团队仍然从目标院校招聘应届毕业生进入内核团队时。我不确定你们团队是否还在这样做，所以我用了过去时。
问题：
* 新员工要花多少时间进行培训？你们会给他们例如 3 个月的时间来赶上进度并开始处理小任务，还是采用“适者生存”的策略？
* 如果有结构化培训，你们使用什么材料？主要是阅读内部文档，还是跟随资深员工，或者阅读关于内核的已出版书籍，还是其他什么？
* 培训如何解决代码质量问题？它会为新员工提供什么样的练习/材料来处理代码质量问题？在内核团队中，代码质量意味着什么？
* 培训如何解决性能问题？你们有多重视内存占用？
* 培训是否使用生产操作系统的最小版本，并要求新员工复制现有功能？
* 你们首先向新员工介绍哪些“子系统”/“模块”（例如，vm/boot/driver/scheduler/fs/etc.）？
* 培训如何解决内部工具问题，尤其是内核调试？
感谢阅读，希望我能得到一些答案 :D
查看原文
Context: I'm wondering how did OS kernel teams provide training to their new hires.<p>I'm mostly curious about:<p>- NT kernel / XNU kernel / Android kernel teams, or anything that is not OSS like Linux, e.g. some prop RTOS team would also count<p>- When teams still hired new graduates from targeted schools into kernel teams. Not sure whether your team is still doing this so I put up a past tense.<p>Questions:<p>- How much time do new hires work on the training? Do you give e.g. 3 months to catch up and start to work on small tickets, or it's a swim or die strategy?<p>- If there is any structured training, what material do you use? Is it mostly reading internal docs, or shadowing seniors, or published books about the kernel, or something else?<p>- How does the training address code quality? What kind of exercises/material does it lay on the new hires to work on code quality? What does code quality mean in the kernel team?<p>- How does the training address performance? How much do you care about memory footprint?<p>- Does the training use a minimal version of the production OS and ask the new hires to replicate existing features?<p>- What "subsystems"/"modules" (e.g. vm/boot/driver/scheduler/fs/etc.) do you expose to the new hires first?<p>- How does the training address internal tooling, especially kernel debugging?<p>Thanks for reading, and hopefully I get some answers :D