I have worked as an interviewer in FB and interviewed more than 200 people, so I am quite familiar with FLG’s interview mode.
Many others only answered the first question, followed by a series of questions.
Domestic fresh graduates are not familiar with Leetcode brush can enter Google?
We all agree on this, but LET me just say why not from different angles.
Because most of the interview questions from Google you can’t find in LeetCode/LintCode.
One of the characteristics of Google’s algorithm interview is that it likes new or variant questions, you can go to the book, and there are very few original questions. So LC brush more skilled, no real understanding of the words, may change the vest will not do.
In addition, in the algorithm interview, if you fail a question quickly, the interviewer may ask another question or follow up. It’s no use memorizing answers without thinking.
Everyone is about the same level, so what is the final standard to distinguish?
At big companies like Google, the interview evaluation system is already standard. We’re only talking about algorithmic interviews here, but actually system design and behavior problem interviews.
Let’s look at a problem and see how it feels with an example
Longest Palindromic Substring
For this question, interview comments from high to low are as follows
Strong Hire
Use O(N) or O(Nlogn) Algorithm to achieve (Manacher’s Algorithm or Suffix Array), and the code is excellent, bug-free or have a small Bug, but can find and solve their own, without too much prompting
Hire
Time complexity O(n^2) algorithm can be implemented using enumeration method and dynamic programming respectively. And the code quality qualified, no Bug, no duplicate code, without the interviewer to prompt.
Weak Hire
It only uses one of the O(n^2) algorithms, and the code is pretty good, with a few bugs, and the interviewer can give you some hints.
No Hire
Can only come up with one O(n^2) algorithm, but it’s buggy, or it needs a lot of hints.
Strong No Hire
I can’t even think of an O(n^2) algorithm
Four or five rounds of interviews
Yes >= 1 Strong No Hire => No offer Yes >= 2 No Hire => No offer 1 No Hire + 1 Weak Hire => No offer 1 No offer A Weak Hire = a Weak Hire = a Weak Hire
At one extreme: a Strong Hire + a Strong No Hire => set up a meeting to discuss it, usually resulting in an Offer or No Offer.
To sum up, ** interviews don’t necessarily require you to use optimal complexity algorithms to solve problems. ** While only using LC, it is easy to make the mistake of using the best solution to do the problem.
And the code is not just written, code quality is also important.
Good code quality includes: Bug free, good Coding Style (including variable names with meaningful naming conventions, reasonable use of Spaces, good use of blank lines, etc.), easy to read logic, no redundant code, boundary detection and exception handling.
Is it easy to intern and then get a job?
It depends on the company, unfortunately, it is not easy to become a full-time intern. ** At least in North America, domestic competition should be greater, not to mention full-time, internship is hard to get.
Tell us what it takes to become a Google intern in North America.
Four feedback feedback are required for Google internship to become a regular internship. One feedback can be collected from host and co-host for each internship.
There are two ways to get the remaining two: another internship or two rounds of interviews.
So if you want to get two Google internships, prepare at least a year in advance. And the performance of the first internship determines whether an offer will be given to the second internship. Most people only have enough time for one internship and two rounds of interviews before they get a job offer.
About all aspects of the algorithm interview, I talked a lot in the “nine chapter algorithm class”, in addition to how to correctly and efficiently brush questions, and how to communicate with the interviewer, how to improve the quality of the code, the interview style of the big factory and so on are involved. In the latest recruitment, there are 3 free experience classes, you can receive nine chapters algorithm selected 100 algorithm interview questions.