Today, more and more people want to enter the field of artificial intelligence, and the direct path to enter the field is to pursue a career in machine learning.
But how should machine learning practitioners plan their careers? What are the key factors that make for a successful career?
How to find a machine learning job
The competition for machine learning jobs is fierce, but what do recruiting companies think of as their ideal candidate?
These are the abilities we need to have before applying for a job.
Programming skills
People who want to work in machine learning research need to know at least the basics of programming. On average, programmers need to know two to three languages at an intermediate level.
Machine learning practitioners need programming skills and an understanding of many commonly used programming languages, such as Python, Java, JavaScript, and R.
Ability to understand technical issues
After passing the resume review stage, the next step is the interview. In both interview stages, machine learning-related questions will be covered.
The interviewer might ask you to explain the differences between gradient descent variants or describe the unique characteristics of a new neural network architecture.
The ability to practically apply theoretical knowledge
Before the interview, you may have spent several years in an academic institution and used the university’s theory-based assessment method. Recruiting companies want ideal candidates who understand machine learning theory and can apply concepts, techniques, and ideas.
For example, you first need to understand the benefits of batch normalization conceptually, but to stand out from the competition, you have to learn to use Jupyter notebook or GitHub repo, have a certain amount of work, and the accompanying portfolio to demonstrate the technology, ideas, and solutions. A great way to ask questions.
Ability to learn continuously
In machine learning, new technologies emerge every day, new tools and libraries emerge one after another, and many research papers are published monthly. To enter machine learning, you need to be able to learn continuously. In deep learning, new neural networks often emerge that achieves state-of-the-art performance on specific computer vision tasks. Recruiting companies hope employees are dissatisfied with their current positions and constantly pursuing growth. Machine learning practitioners are often at the forefront of emerging technologies in the AI ​​industry.
Successful AI Practitioner Model
AI covers many subfields, such as machine learning, natural language processing, speech recognition, neural networks, computer vision, image processing, and more. Successful AI practitioners should take a “T”-shaped approach when learning AI subfields. That is, broadly learn the basics of most subfields of AI and then dive deep into specific areas of expertise.
professional knowledge
The composition of professional knowledge is as follows:
- project
- open source contribution
- Research
- cooperate
Again, working on a personal project in a specific area will deepen your knowledge and expertise, making you a successful AI practitioner or even a more hands-on role.
How to choose a job
the market needs individuals with machine learning expertise. It allows people with machine learning knowledge and skills to have more choices when applying for jobs, but it is also inevitable to make wrong decisions. Here are some pieces of advice on choosing a job and having a happy and meaningful career.
work with a good team
Choosing an excellent team is very important. When selecting a team, the following factors need to be considered:
- influences
- communicate
- growing up
Andrew Ng suggested that one should work in a team that can easily interact with team members. Generally, such a team contains 10-30 people. He recommends paying attention to how hard your team is working and whether the personalities and work ethic of team members positively impact you.
Individuals on a team are the ones you spend most of your time with. According to behavioral science research, your competence is ultimately the average of the five people you spend the most time with.
Know your position
Make sure you understand the job before deciding to accept it. Often, job descriptions on job postings need to reflect the nature of the functions and responsibilities required for the actual job. Sometimes, responsibilities at work are greatly exaggerated, leading to disappointment.
Sometimes, recruiting firms need to pay more attention to the expected workload of new hires, which can lead to premature career burnout.
The best way to avoid disappointment and burnout is to talk to your direct manager and understand expectations for tasks and delivery times. Also, talk to team members in similar positions and ask questions about their day-to-day activities.
Ignore the company itself
A company’s brand is usually the image a company conveys itself to the outside world. Companies usually only show their best side to the outside world, making you slightly biased when choosing a company.
From a general experience, the correlation between a company’s brand and your personal experience with the company is weak.
When choosing an AI position, it is more important to consider the team than the company, and this conclusion applies to almost any industry job.
For example, an image classification & machine learning project for an oil company and a medical center differ only in the dataset used to train the machine learning model and its application. Machine learning skills are transferable across industries.
Keep looking forward to job opportunities and consider long-term goals
Humans are motivated creatures both internally and externally, and when both are lacking, we start to regret, worry, and sometimes even slip into a state of depression.
It would be best if you chose a position that best suits your personal goals and long-term development.
Misunderstandings that need to be avoided in a machine learning career
Become an AI “jack of all trades.”
You don’t have to be a “jack of all trades” to work in AI, and having a shallow knowledge of all subfields of AI does not guarantee success in a career in AI or machine learning. Specializing in a field pays off in the long run, especially in AI, where specialization is necessary.
impatient
In a rapidly evolving field such as machine learning, we almost always feel that we need to absorb information faster. In any subfield of AI, it takes time to build fundamental skills and expertise.
no substantive items
Project experience or a portfolio has many benefits for developing a machine learning career. However, your portfolio must be reproducible enough to impress the interviewer.
A lot of energy and time is required to create a work with sufficient impact. Ten mediocre projects are not as good as 2-3 impactful projects, quality over quantity.
conclusion
- work in a conducive environment that facilitates learning;
- Doing projects with real meaning, that is, doing business to improve the quality of life of others;
- Make decisions that contribute to your personal goals, setting the stage for long-term success.