我面试了参与制定 IPv6协议，TCP协议的44岁的印度程序员
很久没写公众号文章了，今天分享一下有意思的经历。
这位大佬45岁了，基本都是95年左右，10年里都在Intel里工作，参与制定了这些协议。
Co-author of RFC 3964 - Security considerations for 6to4 Co-author of RFC 4389 - Neighbor Discovery Proxies (ND Proxy) US Patents #6865591, #7502823 - Distributed fault-tolerant applications
后来他就去当各种CEO，CTO了，一直在印度，最近才来面我打工的厂。
当时面试有点急，我来不及看他的简历，后来看到这一串RFC协议，我不太敢相信一个参与过制定协议的人只是来面这个职位，职位没有特别高，跟我密切合作的一个同事也有这个职级。
为了确认，我用他名字确认了一下，确实他是有参与，感谢列表里也有他名字。
我就问了下他的经历，刚好我了解TCP的RFC（请看此文章：知乎万赞：看完这篇再问三次握手就教面试官做人），在各自自我介绍，一番寒暄后， 就用此开始了面试。
以下是我的一面面试反馈，就不翻译成中文了，而且这个只是我的反馈并不是官方的，只是内部分享给另一个人e，而且他肯定过，公开也没啥问题。
Understanding of RFC
I ask his understanding of TCP as he has worked on some RFC of IPv4, Proxy etc.The first question is why a three-way handshake is necessary, and he answers this question by explaining how a three-way handshake is required to sync things and make the protocol reliable. Then I asked him if it was possible to do the two-way handshake only. Instead of giving an exact answer, he explains how a three-way handshake works again.I asked more specifically if A and B use a two-way handshake, is it ok? Are all the msg sent to B reliable? He said yes, and A could send all the msg to B. I asked more specifically whether he knew there is a sequence number in TCP protocol, as the "exact and correct" answer of TCP three-way handshake should be explained with this. His explanation and usage sequence number proved he understands how to design a protocol and why TCP is a reliable protocol.
Algorithm
Then I ask the algorithm question: https://leetcode.com/problems/longest-substring-without-repeating-characters/description/
His solution with recursive is quick to come up. His implementation of the IDE is rapid too.
After his code passed all the tests he created, I asked him about the time complexity of the recursive solution. He explained this by expanding the case "hello" and said On^2.
精简了一些内容， 期间还有shadow的面试官问了其他问题，我就不分享了。
至于系统设计，管理能力，经验方面的，就是二面三面的事情了，就此打住。
总的来说，尽管当了一系列CTO、CEO，面外企engineer或者manager，除了特别高位，不然还是得按游戏规则玩，那就是LeetCode。
看得出他还是准备了一番，解题挺快。我发现他在面试时倾向于多说，于是面试官就没啥时间问了。。好处就是，这样子就能更少地暴露出问题跟缺点，而且拖到时间够了面试就结束了。
把主动权握在自己手里，而不是让面试官一直问问问。
如果他当年没有一直创业，眼光好一些，现在应该也挺财富自由了。嘛，选择了创业去赌，现在这样子也不差吧。
好久没写中文文章了，之前一直写英文的RFC，现在写东西都想带个which跟that还有其他修饰了。
如果喜欢这篇文章，麻烦大家点赞、好看、转发三连，拜托了，这对我很重要。（顺便收藏下公众号，下次就可以立刻访问了）
