S**I
发帖数: 15689 | 1 ☆─────────────────────────────────────☆
gzou (gzou) 于 (Thu May 12 02:26:35 2011, 美东) 提到:
马上就要G on site了，
求祝福。
下面是从本版收集到的Google的试题，便于大家查询。
申明：有的附带有解释说明的，也来自于本版或者网络，大家自己看， 不保证真确
http://www.mitbbs.com/article_t1/JobHunting/31847453_0_1.html
本人ECE fresh PhD，背景是电路/EDA，跟G业务基本没什么关系
同学内部推荐的，很简单的一次电面就给了onsite
题都不难，但是自己没把握好机会，出了一些小bug。
总的感觉，出错就是硬伤，宁可从最简单的算法写起，也不能出错。
电面：
1，Skip list， http://en.wikipedia.org/wiki/Skip_list
写code实现struct skip_list * find(struct skip_list *head, int value)
2，sorted array... 阅读全帖 |
|
S**I
发帖数: 15689 | 2 ☆─────────────────────────────────────☆
gzou (gzou) 于 (Thu May 12 02:26:35 2011, 美东) 提到:
马上就要G on site了，
求祝福。
下面是从本版收集到的Google的试题，便于大家查询。
申明：有的附带有解释说明的，也来自于本版或者网络，大家自己看， 不保证真确
http://www.mitbbs.com/article_t1/JobHunting/31847453_0_1.html
本人ECE fresh PhD，背景是电路/EDA，跟G业务基本没什么关系
同学内部推荐的，很简单的一次电面就给了onsite
题都不难，但是自己没把握好机会，出了一些小bug。
总的感觉，出错就是硬伤，宁可从最简单的算法写起，也不能出错。
电面：
1，Skip list， http://en.wikipedia.org/wiki/Skip_list
写code实现struct skip_list * find(struct skip_list *head, int value)
2，sorted array... 阅读全帖 |
|
p*******e
发帖数: 37 | 3 The interview scheduling process was by far the worst in all my professional
career. My phone interviews got cancelled and rescheduled twice! During one
of the cancelled interviews, I was not notified the cancellation until 3
hours after the scheduled interview time!! Of course I was apologized for at
both events. But those apologies did not change my impression of LinkedIn:
the company was not mature, nor professional. I almost decided to take back
my application to LinkedIn.
During the onsite ... 阅读全帖 |
|
b*****t
发帖数: 72 | 4 刚电面完BB，写点面经：
1. static in C and C++
2. public inheritance vs private inheritance
3. four default function allocated by the compiler
4. shallow copy vs deep copy
5. exception mechanism
大概只面了20多分钟，基本都答出来了，但是表述不太清楚，还需要多练练，不知道有
没有好的建议？
希望能有on site. Bless |
|
h*****3
发帖数: 1391 | 5 只谈c++，不谈其他，比如stl，我自个觉得属于data structure的。当然也有一些我认
为很有可能被面试到的概念题。想到哪写到哪吧。很惭愧，大家提的那3本书都没看过
。也许 C++ PRIMER看过。
1) 为啥要用c++;
2) encapsulate, polymophysim,inheritance的概念；
3）class的概念
4）default constructor是啥样的；
5) 写copy constructor要注意啥
6) 写 operator assignment的时候要注意啥
7）can constructor be virtual? why it can or can not? how about destructor.
8) what's the default functions of class
9) describe virtual table
10) what's the size of empty class
11) what's the difference of class and struct
12) what's the dif... 阅读全帖 |
|
t*****e
发帖数: 344 | 6 I don't think your c skill mattered at all. it's obvious that they wanted
to see how you behave outside your comfort zone and you failed.
the first two question were meant to see how familiar you are with the
stuffs you touched.
the question about writing code in c is to see how much you really know
and how you can adapt from what you know to what you are are not sure.
pure technically, its weird that you know pointer but don't about deep/
shallow copy. secondly, you should have mentioned st... 阅读全帖 |
|
t*****e
发帖数: 344 | 7 I don't think your c skill mattered at all. it's obvious that they wanted
to see how you behave outside your comfort zone and you failed.
the first two question were meant to see how familiar you are with the
stuffs you touched.
the question about writing code in c is to see how much you really know
and how you can adapt from what you know to what you are are not sure.
pure technically, its weird that you know pointer but don't about deep/
shallow copy. secondly, you should have mentioned st... 阅读全帖 |
|
w**********o
发帖数: 140 | 8 以前看的， 發現保存過， 發上來福利下各位。
由於CSS沒保存， 感覺有些shallow， 希望各位多多包涵。
不解釋：http://bit.ly/12qG8Xy
順便求推薦， 麻煩MSG我！ |
|
|
a***y
发帖数: 852 | 10 顶这个，学术圈的state-of-the-art research和工业界的de-facto还是不一样的
但是目的本身也一样，学术界本质目的还是求新知。work的好的但是已经被充分理解的
，或者heuristic没有太大通用意义的发不出来也是正常
classification算法方面我觉得random forest, deep learning, boosting相关的都比
SVM更实用。SVM主要是背后的learning theory牛逼，算法本身已经有点过时了，因为
复杂度高并且本质上是shallow learning，而且不容易fine tune，但是理论不会过时
，因为理论就算暂时解释不了实践，也还是可以持续发展的。
clustering目前无解，因为问题本身定义是模糊的，对任意数据最多能够假设一个
gaussian mixture，也就是用k-means。很多文章也在质疑这个是science 还是 art。
但是可以期待一个好算法帮助选择k-means里面的k，同时又像kmeans本身一样高效。
Bayesian topic modeling可以做这个但感觉没有太大前途。未来... 阅读全帖 |
|
a***y
发帖数: 852 | 11 顶这个，学术圈的state-of-the-art research和工业界的de-facto还是不一样的
但是目的本身也一样，学术界本质目的还是求新知。work的好的但是已经被充分理解的
，或者heuristic没有太大通用意义的发不出来也是正常
classification算法方面我觉得random forest, deep learning, boosting相关的都比
SVM更实用。SVM主要是背后的learning theory牛逼，算法本身已经有点过时了，因为
复杂度高并且本质上是shallow learning，而且不容易fine tune，但是理论不会过时
，因为理论就算暂时解释不了实践，也还是可以持续发展的。
clustering目前无解，因为问题本身定义是模糊的，对任意数据最多能够假设一个
gaussian mixture，也就是用k-means。很多文章也在质疑这个是science 还是 art。
但是可以期待一个好算法帮助选择k-means里面的k，同时又像kmeans本身一样高效。
Bayesian topic modeling可以做这个但感觉没有太大前途。未来... 阅读全帖 |
|
G******n
发帖数: 572 | 12 娃满月了，看着他对这个世界的好奇，我无法入睡。
物理phd最后一年，早已对学术界失望，对学术生涯失去热情。老板很supportive，今
年之内完成论文。之前三天打鱼准备了一会儿编程，面了几个，都没有结果。可能是之
前太多事情让自己分心，而且比较迷茫，没有一个准确的目标。（现在也不确定。。。）
现在生活慢慢稳定下来，虽然赶不上summer intern了，还是希望能在毕业之前找到工
作。
地点：大波士顿地区（至少暂时不能离开），纽约地区也可以试试
目标：entry-level 大公司码工，quant，data scientist （支持H1b和opt extension
的）
背景：top20 Physics phd，本科国内top3，成绩和学习能力没问题，数学和解题能力
不错，但没有实习经历。phd方向实验和理论，显微镜成像图像处理，数据分析，MC数
值模拟
技能：C++很熟练，Matlab。其他的都会一些，core java/j2ee, machine learning,
sql都有听过公开课和做作业，数据结构和算法会的有限。leetcode做（看）过20%，没
有认真去刷
差不多... 阅读全帖 |
|
r*******i
发帖数: 43 | 13 来自主题: JobHunting版 - KCG面经 刚电面过KCG,C++ developer职位，应该是不会继续了。
全都是C++问题。
1.空对象的长度是多少？
解释一下什么是deep copy,什么是shallow copy
2.实现copy constructure, operator=，别忘了要判断是不是自己。
3.vector的特点是什么。
4.vector erase是怎么记录剩余的数据项的。我不知道vector的erase是具体怎么做的
，我跟他说用hashmap来记录剩余的项，这样每次删除是O(logn)(但index不是变logn了
?在重整后变回O(1))。看了下stl的说明，原来就是每次erase，把后面的数据挪动一遍。
vector 怎么做到index,insert,erase都是O(1)？请教各位。谢谢。 |
|
U**m
发帖数: 313 | 14 背景：EE PhD signal/image processing方向，6月毕业，跟老板是cs系的，所以也做
了computer vision和machine learning的研究。没有实习经历（硬伤阿硬伤）。
一直有点悠哉游哉，到了3月底4月的时候才开始一边写论文准备答辩一边投简历。
leetcode刷了100多题没刷完，career up大概从头到尾简略看了一下。面试题目我就讲
一下大概哪个方面。
简历准备： 在校的同学不妨考虑学校的career center。一般都有很多辅导的讲座，也
帮助大家改简历。还有就是已经工作的，最好是有面试别人机会的师兄师姐，联系一下
让人帮忙瞧瞧。另外版上精化文章要尽量学习。
感觉简历还是需要稍微准备一下，至少能够过简历关。
投简历：无非就是网上海投，请人refer，以及直接联系HR。
一local小公司：
学校的career center投的简历，过了两三天HM直接联系（小公司没有HR-_-）。
phone interview问的是基本问题，类似于what is the difference between "deep"
copy and "sh... 阅读全帖 |
|
c***z
发帖数: 6348 | 15 【 以下文字转载自 Working 讨论区 】
发信人: chaoz (晨钟暮鼓), 信区: Working
标 题: transend into expertship
发信站: BBS 未名空间站 (Fri Dec 12 18:24:31 2014, 美东)
I feel I am in the stage and the following excerpt describes it so well that
I would like to share here:
From the book "Win without Pitching"
http://www.winwithoutpitching.com/read-it-online/the-passion-di
Selectivity Deepens with Expertise
As our expertise deepens, so too does our ability to be selective. Expertise
forces selectivity.
The generalist is drawn to the... 阅读全帖 |
|
E***e
发帖数: 3430 | 16 You think you know everything simply because you actually know nothing.
Being arrogant but shallow is the key reason why Ah-San are taking such a
ride on you Chinese code farmers.
At least in my area I can easily knock all Ah-San out in phone interviews
within three questions, but you have to enjoy your curry offices
LOL |
|
x*******o
发帖数: 74 | 17 这就开始英文了。。。
楼主上来问银行modeling的活，你上来就“物理经济统计对银行用处都不大吧”，你多
大能耐去否定一群人
这年头搞跨学科研究太平常了，有学习方法技能的去学就是了，还真觉得你这东西那么
难搞么
“Being arrogant but shallow”用来形容你最合适不过了，不超俩句话就开始人身攻
击和扣帽子，看来你工作环境也挺紧张。
面试灭老印值得赞扬，继续努力，就是别太arrogant把国人都给全灭了 |
|
a**********0
发帖数: 422 | 18 如果是shallow clone 一个新obj 是不是比用constructor新产生一个 obj更快呢 |
|
j******o
发帖数: 4219 | 19 shallow copy一般都比较快啊，因为只cpoy reference |
|
d********o
发帖数: 153 | 20 我在以前的工作中和亚特兰大的人有过接触，还都应该算是成功的小业主（美国人，其
中有一个说是亚特兰大有一条街的名字就是以他们家族的名字命名的）。但是这个接触
经历不是太愉快的。感觉很shallow,那自我感觉简直就太好了，好像这世界上，除了亚
特兰大，就都属于穷山恶水。而且在工作的态度中，你也处处感觉到他们的那种高人一
等的优越感。
对于搬家到亚特兰大，我们的顾虑就是孩子的教育（GA教育差是世界人都知道的），但
像我上面说的，差区也有好学校。另外对于亚特兰大人的这种莫名其妙的优越感（对我
而言，是一种浅薄），我其实也有点担心。但任何事情都没有十全十美，所以就看你更
看重哪些了。
说到学校的排名，newsweek的排名还是一个大家比较看重的依据。不管它是按照什么依
据，除非你有朋友真正经历过这个学校的教育，还有如果不嫌麻烦的话，打电话到学校
问往届学生的去向统计数据，也大致可以看出个所以然。 |
|
H******7
发帖数: 34403 | 21 有空读读这个,内有详细说明
http://ucanr.org/freepubs/docs/8044.pdf
Water as infrequently as possible.
Thoroughly water when you do water so moisture soaks down to the roots.
Exceptions to this general rule would be for newly seeded lawns where the
surface needs to stay moist, newly sodded lawns that have not yet rooted
into the soil, or when summer patch disease is a problem (see Lawn
Diseases). Otherwise, avoid frequent waterings that promote shallower root
systems and weeds (e.g., crabgrass).
to |
|
g*****9
发帖数: 4125 | 22 You need to water more, time is not a good guage, you need about 1 inch each
time, you need translate that into time.
Twice a week with only 15 min each will cause the root system to be very
shallow, if you can not water every day, water every other day for now until
better.
Your grass love hot weather and full sun, it will not do well in shade/
partial shade area. Your first picture is showing your house is blocking the
sun, but I am not sure when its taken. |
|
s*******i
发帖数: 741 | 23 maple. Oak has shallow root. |
|
O**J
发帖数: 475 | 24 maybe initially, but once the root starts to get deep into
the ground, you don't need to. think in nature, would it
rain everyday?
when watering, may sure water soaks deep into the ground,
that's how the root would grow down. If you just water a
little bit, the root will stay shallow and dries up when
you don't water. |
|
p*******p
发帖数: 220 | 25 Bunch of scratches are created on my the window glass. The scratches are
very thin and shallow like hairs.
Anyone has experience or good idea to remove them?
The scratches really bother me.Many thanks in advance!!! |
|
s**x
发帖数: 7506 | 26 什么样的好啊？偶需要 shallow depth 的， 疑问橱柜不大。 有经验的指点一下？
多谢！ |
|
j***b
发帖数: 5901 | 27 饮用水必须打井到第一个隔水岩层以下的蓄水岩层中，浅不了。开销感觉不会少。
其实用水更多是灌溉吧，室内用水再多能用多少？有些地方，比如听说florida，适合
打shallow well用于灌溉，20 feet左右就行了。开销要小得多。 |
|
j***b
发帖数: 5901 | 28 You need to contact the right government department (which I don't know, but
you can try getting information from city hall), ask them about local
geology and regulation. Mostly, to get water from shallow well, you need:
1. high water table.
2. sandy soil.
3. It's allowed by the city.
Your local government should have this information. |
|
w*******g
发帖数: 9932 | 29 sink is too shallow. dont like separate mirrors and carbinet on the sides |
|
B******N
发帖数: 445 | 30 U pin straighten.
The trick is you push the blocker which is really close to outside surface.
the position is 12 o'clock of the hole. It's really easy to miss.
shallow 12 o'clock is the position. |
|
H******7
发帖数: 34403 | 31 pool in wylie, tx
Amount: $15,000.00
Posted by: bl in wylie, TX. Posted: November 30th, 2010 12:11PM
Material: vinyl-lined Type of Pool: in ground
Can the person who posted this give me the name of the company?
Was this post helpful to you? yes no Report prohibited or spam
In ground custom pool
Amount: $56,000.00
Posted by: a user in Riverview, FL. Posted: March 28th, 2011 09:03AM
Material: Concret and shale Type of Pool: in ground
Sq Ft of Pool: 850
Custom pool with zero en... 阅读全帖 |
|
x*******u
发帖数: 194 | 32 some trees are of the fast growing/shallow root system types, such as beech,
birch trees. it is not my 1st time to see some tall trees falling down
after a severe thunderstorm/tornado to lean against hosuehold items (
including the house and car) and block the traffic
if there are big tree trunk over hanging the house/drive way and sticking
out long, you don't want any dead limbs/branches to fall from the sky.
some old trees may look healthy from outside, but there could be holes
inside due to m... 阅读全帖 |
|
f****i
发帖数: 20252 | 33 oh，frost protected shallow foundation
你家在大平原？ |
|
a*****e
发帖数: 4936 | 34 新房，builder给了个disclouser，说地下水位偏高。那片地都有这个问题，the
subdivision has a high water table which is fairly common in that area。他们
采取了一些措施：
1：crawspace底下高一些，所以整体crawspace里面高度低一些；
2：crawspace里面装泵；
3：crawspace里面做了防水防潮处理。
这样的房子可以考虑吗？builder在当地挺有名的。下面是builder写的：
-The floor of the crawlspaces are shallow and calculated to be above
seasonal high water levels.
-As backup to this there will be a sump pump installed in case of an
excessive wet seasons down the road
-We have also foam insulated the crawlspace wall a... 阅读全帖 |
|
|
|
F***Q
发帖数: 6599 | 37
如果池子太浅就不行。我家的原来有跳板，后来前屋主在93年的时候去掉了。他们走的
时候给我们留的条子说pool is too shallow for diving board。地上现在还有当年
diving board support的印。我的深水区是8 ft。 |
|
|
t****t
发帖数: 610 | 39 Is your trap too shallow so it can't hold enough water? |
|
b**********y
发帖数: 7371 | 40 just the facts
If you categorize ppl with their background such as race and nationality,
then you are demostrating your shallowness |
|
|
a*****n
发帖数: 5158 | 42 1.0 ROOF COVERINGSAttention Required
The roof has pitched cement tile roofing and the balance of roofing area(s)
has a flat/low slope design covered with built up roofing (asphalt and
gravel). Due to the nature of the flat roof construction water will likely
pond on the roof surface after normal rainfall.
The pitched cement tile roof at upper area appears intact and functional
with lower NE pitched cement tile roof has some cracked/broken tiles
observed at middle SW area.
The asphalt and gravel ... 阅读全帖 |
|
f******I
发帖数: 769 | 43
this is your normal REM sleep where you have lots of dreams, breathing is
fast and shallow, heart rate is high, if you are a guy, you probably will
have penile erection(if you are embarrassed, can try to sing national anthem
), for female, clitoral enlargement(have no idea how it feels like),
我额头下巴长痘痘多，跟这个有联系吗？
don't think so, |
|
l*h
发帖数: 4124 | 44 A darker spot may last a few days to a few weeks, but no scars for removal
of such a shallow cyst.
It's a very short procedure done in the office. There is no rush to have it
done either. |
|
l*h
发帖数: 4124 | 45 you need to know what you are saying is very shallow and more important,
quite UNETHICAL. you are banishing what are proven in clinical practice
without evidence.
unfortunately to you, my research part of work is targeted therapy in solid
tumors. I can also tell you that only in a very few situations the factors
predicting response to targeted therapy are found.
depending |
|
l*h
发帖数: 4124 | 46 just judging from this path report, it is early and invasion is still
shallow. |
|
l*h
发帖数: 4124 | 47 it depends on how deep and how clean/dirty the cut is. if it's shallow and
clean, a band-aid is sufficient. |
|
k****o
发帖数: 83 | 48 最新消息：
麻省理工大学MIT正面向全球征召应对气候问题的高科技解决方案。
该活动是一次全球性的技术方案大竞赛，开辟了很多分类组。
提交方案的收稿、初选酝酿期为3个月，半决赛1个月，决赛1个月。3天前截稿期已到，
各组都收到了大量技术方案书。
组委会配备了各方面的专家教授，将严格筛选进入半决赛、总决赛的名单，并为各组各
类获奖者颁发1万美元，并引荐给世界知名大财团供其投资参考，或MIT亲自孵化。围观
的风险投资公司、天使基金等小投资人等，虽无美国麻省理工大学的关照，但仍可根据
自己的兴趣半途利诱“拉客”。
全球的吃瓜群众，也可自由注册帐号后，参与点赞投票评选活动。组委会承认围观者的
评选结果，并颁发“最受大众欢迎奖”。如果群众与专家所见略同的话，专家奖与大众
奖会合一。
我代表千瓦厚实验室参赛，分别在在能源组和交通组提交了方案书：
Proposal 1:
Commercial flapping-shutter ornithopter era will come soon
http://www.climatecolab.org/contests/2017/transportation/c... 阅读全帖 |
|
n***y
发帖数: 15001 | 49 if that is the case...
go walmart...
find those plastic container...
some are big and shallow... |
|
s***l
发帖数: 2236 | 50 You are a shallow, self-centered, ill-mannered mom. sorry for your adorable
twin babies. |
|
