Hello everyone, I went through ups and downs during the summer internship in 2023, and finally went to Baidu to do large-scale model-related work. Although my undergraduates and masters are all computer majors, my own strength is still weak, and my hard strength is not like many people who start preparing early , so the whole process was not very smooth. This article mainly shares my experience with you.
Although I made up my mind at the beginning that I didn’t intend to study for a Ph. D. and went to work, but because the research direction of my laboratory and the skills required for the algorithm post were slightly out of track, I spent most of my time wondering whether I should find an algorithm post. In the end, the large model became popular later, and I just decided to use the all in algorithm because I had the experience of RL self-study and participating in competitions, so here I mainly talk about my personal summer internship experience from scratch, hoping to give and I have some help from fellow students who have had similar experiences.
Technology Exchange
Technology must learn to share and communicate, and it is not recommended to work behind closed doors. One person can go fast, and a group of people can go farther.
Good articles are inseparable from fans’ sharing, communication, and recommendations. Dry data, data sharing, data, and technical exchange improvements can all be obtained by adding an exchange group. The group has more than 2,000 members. The best way to add notes is: source + Interest direction, easy to find like-minded friends.
Method ①, add WeChat account: mlc2060, remarks: from CSDN + add group
method ②, WeChat search official account: machine learning community, background reply: add group
dry goods sharing
Most of the content I share is suitable for students who have no time to prepare for the interview due to other reasons, and want to find an internship and are eager to learn quickly . The content that needs to be learned is compressed into a short period of one or two months . If the interviewer conducts a targeted investigation, it can actually easily determine whether you are memorizing by rote or thoroughly understanding the basic principles of technology. The obvious point is that, If the interviewer realizes that the knowledge you have mastered is based on rote memorization rather than active learning , there is basically no chance of passing this interview.
Dry goods 1: Resume
I think the first thing that needs to be carefully prepared is the knowledge content of relevant methods in your resume project , because most interviews will revolve around project experience, and the content involved may have the basic principles of the method used. Use this The reason for the method, whether you have tried other methods and the reasons for the poor results, and some variants of the current method will be the focus of the interview. So what we need to do is not only simply use a certain tool or a certain type of method, but more importantly, we must know the basics and draw inferences from one instance. There is no shortcut to this part, but it is also used in almost every interview after investing time in learning. important knowledge.
Dry goods 2: Likou
I am a student with weak coding ability. It is difficult to give constructive opinions on this part. Personally, I feel that it is better to think long-term when brushing the questions. It is best to start preparing as early as possible. Algorithm questions are alive. It is difficult to achieve the goal of passing high scores by rote memorization like other knowledge. It is more about cultivating thinking. The suggestions I can give include:
1) Do more popular questions , start from the basic questions, don't rush to eat fat in one go, if you can do some medium-difficulty questions but the basic algorithm questions can't be solved in seconds, even if you do it right, you will still lose points ;
2) Pay attention to time- limited exercises when writing questions . If you don’t pay attention to this part, you may be like me. When you are nervous during the interview, your mind will go blank.
Dry goods three: stereotyped
Although in my opinion, the priority of stereotypes is not as good as the project knowledge in the resume, it does not mean that stereotypes can not be prepared. I personally recommend the book "Hundred Faces of Machine Learning" . Many basic algorithms are introduced in great detail. The stealth skills that can be provided are: key chapters focus on key memories, and other chapters focus on understanding when time is insufficient. . But it is best to know the extent to which the interviewer can recall some technical details when asked, instead of not knowing the input and output of the method after reading it and not reading it. Still the same sentence, you must be very careful about the points that will lead to a lot of points being deducted if you don’t do it well. For a technology that you are not very familiar with, I think that even if you output a sentence of relevant content or opinions, the points you get are far greater than those of the interview. The officials themselves do not understand the scores obtained by this technology.
Experience Sharing
Experience 1: Be more targeted
If you really want to go to a company, you can actually understand his interview process first , and at the same time read more about the interview experience. In addition, the pertinence I want to say is reflected in the fact that, for example, when you are about to participate in the interview of this company, if you plan to review Likou, you can learn about the types of questions the company makes. For example: when interviewing code, you generally need to write tests yourself. Use cases, then there is a high probability that binary tree-related questions will not be examined.
By analogy, you can review the company or business department you really want to go to, and get familiar with the core technologies that the other party may use.
Experience 2: invest more and look for more opportunities
Although the employment situation in recent years is not optimistic, in order to find suitable job opportunities, we still encourage everyone to obtain referral information from various channels . The internal referrals that can be found, but specifically refer to the internal referral information of a specific department in the company. There are many benefits specific to the department, so I won’t go into details one by one. The advice I can give here is to broaden the scope of information acquisition. Channels, a good referral opportunity will save me a lot of detours. For me, the machine learning community group and some recruitment information published on Maimai are good sources of information.
Experience 3: Learn to review and persist
Regarding the replay, it is best to prepare a device to record the entire interview during the interview, and then rate yourself after the interview, and then according to the results of the interview, combine these three elements to understand your interview process. flaws and traits that interviewers generally expect to see. Of course, whether the interview is passed or not has a lot to do with luck and HC margin, so I have to mention another key word here: persistence , although persistence is easy to say, but when you see that all your efforts have been in vain , After a lot of interviews that I thought I should have passed but were rejected for various unexpected reasons, it was really difficult to cheer up and face the next interview. I thought about giving up, but I still gritted my teeth and persisted, so I hope that everyone will not be discouraged when they are at the bottom, especially not to let the unsatisfactory process of finding a summer internship affect their mentality and even daily life, and have a positive and optimistic attitude. Really more than anything else.
thank you
Finally, I am very grateful to the friends in the machine learning community, especially the fourth brother, who provided high-quality interview questions for the mock interview and provided me with valuable interview experience, allowing me to reflect and learn a lot from it. In addition, in order to practice my hands through mock interviews before my formal interview, the brothers worked overtime to cooperate with me to coordinate the interview time, which really touched me. It can be said that without the help and encouragement of the DW brothers, I might have given up a long time ago. Thanks again to DW.