[91 Interview] Microsoft big guy cabbage shares algorithm interview experience

background

Recently learned cabbage got Microsoft of the offer , and ready to take other greater the company's offer. I couldn't wait to contact him, hoping that he himself could accept the interview. So this interview draft was met with everyone.

Cabbage is a person who does things very seriously and meticulously. He is meticulous in his work. Basically, you can rest assured that you leave things to him. Who doesn't like this kind of person? I am very optimistic about him, and I can definitely enter a better company.

The following Q is lucifer and A is cabbage.

Interview

• Q: When did you first come into contact with data structures and algorithms (hereinafter referred to as algorithms)?
• A: It started when I was about to jump to a big factory in 20 years. I didn't use algorithms at all in front-end development, so I didn't learn it.
• Q: When did you come into contact with 91 Tianxue Algorithm (hereinafter referred to as 91 days)? How did you learn it?
• A: Through the official account followed by Lucifer's algorithm repo on GitHub, and then learned through the official account.
• Q: What changes did 91 days bring to you?
• A: From not knowing anything to having a little understanding of the basic data structure, I used to only do easy questions, but now some medium questions can also be done, mainly because I learned how to set templates.
• Q: Is there a moment of "insight" in the process of learning algorithms?
• A: Yes, I suddenly found out that when I saw a new question, I would unconsciously go to the previous routine to apply it, and then found that it really works.
• Q: What algorithm are you good at? Can you share it briefly?
• A: I'm not good at it, but I like template questions. For example, if you want to find permutations and combinations, you can write a dfs first. If you find a substring that meets the conditions, you will enter a sliding window. Can be solved.
• Q: Is there anything you want to share with students who have just entered the algorithm?
• A: The template is particularly useful. Whether it is memorized or not, it is a process of understanding at the same time. The solution of the 91 algorithm can help you deepen your memory and help you understand better. In addition, you can brush all types of questions at one time. For example, you can follow the classification of 91, which will be more efficient.
• Q: Relatively speaking, where do you think 91 days is not good enough? How should it be improved?
• A: You can add a few more questions for the same type of questions to deepen your memory. The transition from easy to difficult is not sure if you can make it smoother. It is not easy to understand some types of questions without a CS background. More background may be needed, such as DP and back track. (Because I have been doing front-end development, thinking is always habitual linear thinking, it is not easy to get around)
• Q: What is the learning material that helped you the most in the process of learning algorithms?
• A: LeetCode's problem solving forum and some videos of YouTube.

lucifer Note: Some leetcode problem solutions and tubing videos are really good, and we also recommend them in the lecture notes of "91 days learning algorithm".

• Q: Would you like to share 91 days with your friends?
• A: Yes.

lucifer reviews

Receive cabbage's suggestion. In the next issue of "91 Tianxue Algorithm" Xifa intends to make several changes.

1. We are currently cooperating with official, hoping to integrate Leetbook's 160ff8bd87dfd0 learning plan and LeetBook . This way everyone learns more seamlessly and more efficiently. There will also be a dedicated exchange group.
2. Add picture , simulation and enumeration questions, this part was self-study before
3. Add sorting chapter
4. A few questions have been added before, and corresponding questions need to be deleted in other chapters. I plan to delete the high-frequency interview questions.
5. Part of the problem solution is not perfect, so I will polish it again.
6. Part of the difficulty gradient has a large span, add a little buffer problem or large explanation for a larger span, and try to keep everyone behind.

In addition, it is no longer recommended that you participate in this issue (fourth issue) of "91 Tianxue Algorithm", you can wait for the next issue. And the next period may be shortened to about one month instead of the current 91 days. This is because the rhythm of one question every day is too slow, which will lead to poor learning effects. And if you think you can’t keep up for 91 days, wouldn’t it be even worse for one month? I think this kind of worry is superfluous, because it has little to do with time. It is more about learning methods and skills and after constantly doing the questions.