The title uses “building” instead of “writing” because some experiences take time to prepare. For example, taking a course takes at least half a year, and longer if there are prerequisites. Competing in a Kaggle machine learning competition can take three to six months, and it’s not always possible to achieve results that are good enough to put on your resume. Deliberately building your resume in advance will give you the best chance of getting interviews when you’re looking for a job.
Goals
People who have worked in machine learning know that the metric is the most important thing for models. A model can only solve problems well if it uses the correct loss function to optimize the correct metric. In this article, a machine learning engineer is someone who uses machine learning methods to solve business problems for companies, rather than someone who proposes new machine learning methods and publishes papers.
What is the goal of a company when hiring a machine learning engineer? It is to find someone who is competent, has good communication skills, and is easy to get along with to complete the task. An entry-level machine learning engineer needs to use machine learning to solve a (at least roughly) defined task, and needs to have a certain level of understanding of machine learning and be able to write code. Based on this, we can know what kind of experience will be noticed on a resume.
Standout Experiences
The more relevant and rare the experience, the more likely you are to stand out from the crowd of resumes. Here are a few experiences that I would be impressed by when I see them.
Related Internship
The relevance of an internship to a full-time position can vary greatly from company to company. I have heard of internships where the intern builds models, which is of course very relevant. But I have also heard of large companies hiring machine learning interns to label data, which is not helpful for finding a full-time job in the future. Whether or not it is relevant comes back to our goal: Is this experience relevant to the core daily work of a machine learning engineer?
Machine Learning Thesis
Being able to write a machine learning thesis under supervision means that you are able to customize existing models to achieve better results on a given problem. Publishing a thesis, especially at a tier 1 conference, means that you have the ability to solve problems better. Just as in our work, we need to solve problems unique to our company, we need to at least be able to use existing models, and preferably be able to customize models to achieve better results within acceptable costs (time, machine, manpower, etc.).
Machine Learning Competitions
Here, machine learning competitions refer to the results obtained by participating in competitions with prizes. Machine learning competitions are a great way to learn, and through competitions, you can learn how to apply classroom theory to more realistic data. However, when participating in competitions, you must really think about and try to improve the model or features to get better results. Simply grabbing the first-place code after the competition and changing two lines to re-upload it will not teach you anything and will not add any value to your resume.
Relevant Experiences
The following experiences are too common or less relevant to the core work of machine learning engineers, so they will add some value. They still add value, so you should still put these experiences on your resume.
Related Courses and Curriculum
Related majors generally refer to electrical engineering, computer science, and information management. The graduate school will have different names, and will look at the core courses and master’s thesis topics. To be honest, when looking for the first job or internship, they will definitely look at your academic background. There are only two situations where they will not look at your academic background: you have worked for many years and have other work experience to prove your ability, or you have outstanding experience beyond your academic background (Kaggle solo team wins gold medal, publishes top conference paper, etc.). In comparison, academic background is the easiest to obtain. If you don’t have an academic background, taking more courses will be helpful.
Programming Competitions
ACM ICPC, Meta hacker cup and other algorithm programming competitions. Good programming skill is a big plus for software engineers, and machine learning engineers are also a kind of software engineer. Of course, good programming skills are a plus, but the core skill is still machine learning.
Irrelevant Experiences
Personal Projects for Practice
Nowadays, it is too easy to find a dataset on the internet, build a model, and run a score. It is very easy to achieve 99% accuracy in handwritten recognition on some public datasets. Basically, you can just find a sample on Github and follow the tutorial to run it. Especially now with ChatGPT, anyone who knows a little bit of programming, or even doesn’t know any machine learning, can do it. Something that everyone can do does not add any value to your resume.
Conclusion
The above are my personal opinions. Every company, every position, and even every hiring manager may look at different aspects. The key point is that the more relevant and rare the experience, the more points it will add. Even if AI becomes more powerful in the future and more work can be done by AI, this point will not change.
A good resume can help you get an interview, but whether or not you pass the interview depends on your communication skills and on-the-spot performance. The recruiter will definitely try everything to make the interview performance positively correlated with the work performance, but due to the limited interview time, the interview cannot test every aspect of the work required.
This is the reality, so “interview” is also a professional skill that needs to be practiced. Finding a good job is so complicated. It takes time to research everything from resume to interview to salary negotiation. A resume is a transcript of your achievements in the past few years. Start with the end in mind. If you want to have a outstanding resume in the future, you can start working towards it now.