A salaried and exponent partner for the engineering career video consultation series
- How to become a machine learning engineer
- The Engineering Manager vs. Individual Contributor Path
- The best programming languages for job hunting (in 2022).
Here is a summary of the first one! To watch the full interview, scroll down the article.
How have careers in machine learning and artificial intelligence changed over time and what is behind this evolution?
It is constantly evolving. There is a clear shift away from machine learning and pure data science to a more holistic approach to roles. As companies collect more and more data, it’s inevitable to try to build predictive models. I think this is a natural result of the age of data collection.
What education, technical knowledge, and software skills do machine learning engineers need?
Success is achieved as a machine learning engineer from different walks of life. Although the role has been around for a while, it still feels new. While we seek diversity in our team, we try to hire people whose strengths will mesh well with existing team members.
In terms of education, a solid understanding of computer science and mathematics is standard. Also, I would say:
- software development experience
- sense for business
- statistical modeling skills
- understanding probability
- does a great job with data management
- an understanding of DevOps is helpful
- how to develop and smoothly deploy models in production.
No one person has the same expertise in all of these. It’s incredibly difficult to find someone who immediately ticks all of these boxes, but I suggest that you at least spend some time developing these skills at a minimum. Then identify what areas really drive you and find a team that needs that energy. Because machine learning is constantly changing, chances are you’ll find a time that fits your skills.
Are there computer science degrees and certifications in data science and machine learning?
This is a common mistake for companies that require a degree in machine learning. why? They are relatively new, so there are not many people with these credentials. Frankly, the demand for engineers with a master’s degree does not match the supply.
Many people come, including me machine learning engineering from numeric fields. I have a PhD in physical sciences. Our team includes people with computer science, traditional software engineering, and math backgrounds. All of them fit into the ML role well. You don’t need a special degree to be successful, it seems like it’s very open to different experiences.
How can you change different engineering fields? What roles might be the easiest?
I’d say traditional software engineers who remember math concepts well are pretty flexible, but probably have the easiest time and fastest path to success. Another extremely useful and valuable transferable skill is software development. It may take a long time to develop this.
So, if you are someone who has, you have a big advantage. It is clear that more and more practicing ML engineers are strengthening their software development skills as they become more experienced. So if you’re a software engineer, the core is a natural.
What do machine learning interviews typically look like?
Generally, they resemble interviews for software developer positions. This usually starts with a few technical interviews. You may meet with a cross-functional stakeholder, someone with whom you would most likely collaborate on a project. This person may be from the revenue or product marketing department and is less technical.
The interview with the hiring manager may be nearing the end. As for the technical parts, they are usually divided into software development and algorithm development. Part of direct machine learning can use mathematical concepts more directly.
What are the biggest career opportunities in the machine learning AI space?
There are so many of them! In terms of modeling, working with textual data and natural language processing (NLP) is very important. If you haven’t heard of Transformers Revolution, it’s a new collection of models that are incredibly effective at understanding language.
For machine learning, MLOps is worth a look. At the intersection of machine learning and DevOps, we are seeing more and more roles in companies. Teams need someone who knows how to effectively plan and execute deployments.
There is also room for generalists. Machine learning skills are highly transferable.
On the subject: In a survey of software engineers on the Hired platform, they identified the hottest technology trends as artificial intelligence, machine learning and big data, with 55.1% of respondents ranking it at number one. (Status of Software Engineers Hired for 2022.)
What are the most important skills to develop as your machine learning career progresses?
Versatility and curiosity! As the field is changing and growing rapidly, keep learning! Learn new modeling techniques, technologies, fundamentals – all of it. Don’t burden yourself by investing too much time in any technology.