现在还可以要求att做price match吗？# PDA - 掌中宝
D*e
1 楼
本人CS Fresh PhD,一般学校,专业机器学习.本人实在是不牛,受益于本版,在此攒人品.
申了Microsoft, Google, LinkedIn, Twitter,eBay,都拿到onsite.去湾区只有三
天,只好放弃T.G家开始说过了hiring committee,但拖到三周多后告诉我挂了.由于过于
自信,本以为会签了,导致没有申到今年的H1B.因此对G家充满怨念.拿到M,L,E的OFFER.
思量之后签了M,RSDEII.
先说我的感想:
1)别老想着做题,起决定作用的还是基本功,思维能力,和状态.我有些朋友横扫各大
公司的,他们都不屑于搜面试题来做.而且总有做不到的题,面试时候的发挥很重要;
2)尽管如此,尽量多的去做些题.重复率还是蛮高的;
3)找工作是不确定性蛮大的事情,保持好的心态,自信.
Twitter：
1) Find the median on N machines;
2) Stream sampling;
3) How to evaluate a classification algorithm; How to compare two
algorithms;
4) How to maintain the IP which visits Twitter;
5) Describe Learning to rank;
Google:
本不该说出来，但是他们不仁在先，导致哥没赶上H1B，就别怪哥不义了．何况我
这辈子估计都不会去他们家，无所谓了．
1) Implement a simple calculator (+,-,*,/);
2) Implement "+1" for a large integer;
3) How to match Ads to users;
4) How to extract useful information from a forum webpage (list all
kinds of useful signal you can think of)
5) How to detect the duplicate HTML pages (large scale);
6) Find all the paths between two places on Google map;
7) Find the minimum window in a string which contains all the given
characters;
8) How to debug a random algorithm, i.e., for the same input, the
output
can be different;
9) Inverse a stack give three member functions (forgot details);
prove
why your algorithm is correct;
10) Given the mask of ISP and a huge volume of IPs, find out the top
1000 ISP which contains the maximum number of IPs;
... 其它忘了.
LINKEDIN
从整个面试过程而言，对他们印象很好，很专业。公司定位精准，赢利模式清晰，
是家不错的公司。在股票方面很慷慨。
电面一：CMU读了10年的PHD，问了coding和machine learning。online写了young
matrix lookup, binary search，描述了SVM。
电面二：coding是实现sqrt。同样问ML，甚至涉及到了VC-dimension。
onsite: 安排的见11个人，有些忙的没出现，但还是六轮每轮一个小时整。coding
有neat print, array merge, find common friends on linkedin, producer&
consumer.其它不记得了。
他们家onsite最大的特点是，会大量的问跟他们日常工作相关的东西，非常简单粗
暴。比如：a) how to implement "people you may know";b)how to present the
update streaming of linkedin to users;c) how to match job ads to proper
users; d) how to design the search engine for linkedin; e) how to
organize
news today.
EBAY
电面一：biased coin toss，find the point nearest to a given set of
points.
电面二：a)in a sorted array where elements can be redundant, find the
starting and ending positions for a query int; b)a line of coins
(different value), two ppl pick. He can only pick from one end. Does the
one who picks first can surely win? What's the optimal solution for him?
c)young matrix; d) SVM and SMO.
onsite:也是六轮，每轮近一个小时。有两轮比较tough，包括一个阿三。有点难度
的问题包括：a)closed form solution of linear regression, implement in
matlab; b)design regression algorithms when loss has different weight
for different output (some loss is significant); c)what's the goal of
ebay search engine (this is a tough problem, it's not like web search at
all); d)我想不起来了。。。
最后一轮是见director，他人很好。上来先问我感觉怎么样，我说能过。聊了很久
关于ebay, amazon, taobao之后，问两个问题。第一数组random shuffle。第二个是一张桌
子两个人轮流往上放硬币，放的时候不能跟已经放的重叠。谁放最后一个硬币谁赢，问先放的人有没
有必胜策略。
这位Director人很好。他帮我拎了个很重的包走很久,然后他说顺路，可以开车从San Jose
送我到Sunnyvale。。。还是打车回了。深受感动,一度考虑过签他们.
既然是M的人了,对M家好点,就不说他们的题了,我面的来说他们家的最难,只能说运气的问题
吧. 十月份后M家大涨工资,目前的PACKAGE可以说比哪家都不差,而且给了我20K signing
bonus.
GOOD LUCK TO EVERYONE!
申了Microsoft, Google, LinkedIn, Twitter,eBay,都拿到onsite.去湾区只有三
天,只好放弃T.G家开始说过了hiring committee,但拖到三周多后告诉我挂了.由于过于
自信,本以为会签了,导致没有申到今年的H1B.因此对G家充满怨念.拿到M,L,E的OFFER.
思量之后签了M,RSDEII.
先说我的感想:
1)别老想着做题,起决定作用的还是基本功,思维能力,和状态.我有些朋友横扫各大
公司的,他们都不屑于搜面试题来做.而且总有做不到的题,面试时候的发挥很重要;
2)尽管如此,尽量多的去做些题.重复率还是蛮高的;
3)找工作是不确定性蛮大的事情,保持好的心态,自信.
Twitter：
1) Find the median on N machines;
2) Stream sampling;
3) How to evaluate a classification algorithm; How to compare two
algorithms;
4) How to maintain the IP which visits Twitter;
5) Describe Learning to rank;
Google:
本不该说出来，但是他们不仁在先，导致哥没赶上H1B，就别怪哥不义了．何况我
这辈子估计都不会去他们家，无所谓了．
1) Implement a simple calculator (+,-,*,/);
2) Implement "+1" for a large integer;
3) How to match Ads to users;
4) How to extract useful information from a forum webpage (list all
kinds of useful signal you can think of)
5) How to detect the duplicate HTML pages (large scale);
6) Find all the paths between two places on Google map;
7) Find the minimum window in a string which contains all the given
characters;
8) How to debug a random algorithm, i.e., for the same input, the
output
can be different;
9) Inverse a stack give three member functions (forgot details);
prove
why your algorithm is correct;
10) Given the mask of ISP and a huge volume of IPs, find out the top
1000 ISP which contains the maximum number of IPs;
... 其它忘了.
从整个面试过程而言，对他们印象很好，很专业。公司定位精准，赢利模式清晰，
是家不错的公司。在股票方面很慷慨。
电面一：CMU读了10年的PHD，问了coding和machine learning。online写了young
matrix lookup, binary search，描述了SVM。
电面二：coding是实现sqrt。同样问ML，甚至涉及到了VC-dimension。
onsite: 安排的见11个人，有些忙的没出现，但还是六轮每轮一个小时整。coding
有neat print, array merge, find common friends on linkedin, producer&
consumer.其它不记得了。
他们家onsite最大的特点是，会大量的问跟他们日常工作相关的东西，非常简单粗
暴。比如：a) how to implement "people you may know";b)how to present the
update streaming of linkedin to users;c) how to match job ads to proper
users; d) how to design the search engine for linkedin; e) how to
organize
news today.
EBAY
电面一：biased coin toss，find the point nearest to a given set of
points.
电面二：a)in a sorted array where elements can be redundant, find the
starting and ending positions for a query int; b)a line of coins
(different value), two ppl pick. He can only pick from one end. Does the
one who picks first can surely win? What's the optimal solution for him?
c)young matrix; d) SVM and SMO.
onsite:也是六轮，每轮近一个小时。有两轮比较tough，包括一个阿三。有点难度
的问题包括：a)closed form solution of linear regression, implement in
matlab; b)design regression algorithms when loss has different weight
for different output (some loss is significant); c)what's the goal of
ebay search engine (this is a tough problem, it's not like web search at
all); d)我想不起来了。。。
最后一轮是见director，他人很好。上来先问我感觉怎么样，我说能过。聊了很久
关于ebay, amazon, taobao之后，问两个问题。第一数组random shuffle。第二个是一张桌
子两个人轮流往上放硬币，放的时候不能跟已经放的重叠。谁放最后一个硬币谁赢，问先放的人有没
有必胜策略。
这位Director人很好。他帮我拎了个很重的包走很久,然后他说顺路，可以开车从San Jose
送我到Sunnyvale。。。还是打车回了。深受感动,一度考虑过签他们.
既然是M的人了,对M家好点,就不说他们的题了,我面的来说他们家的最难,只能说运气的问题
吧. 十月份后M家大涨工资,目前的PACKAGE可以说比哪家都不差,而且给了我20K signing
bonus.
GOOD LUCK TO EVERYONE!
