If your company is not doing ML, it (probably) won’t start any time soon either

A recurrent struggle among ML practitioners such as Data Scientists and ML Engineers is that, despite job descriptions and conversations during the interview process, after some time at their job they find themselves with the feeling of doing tons of different things except what they were most excited about - building some ML models.

I’ve been in that situation myself, and have heard it over and over again from friends, colleagues and students. The solution? “Change jobs!” you might think, but the reality is that this is often not the best solution.

In this article, I would like to share my point of view on why this is so common and use it to illustrate how you can make a better decision by having a wider perspective. And, if after all you still decide to go out there and find a new job (I’m not suggesting either option is right), this could help you identify good alternatives more easily.

Why is my company not investing more in ML?

If you’re a professional with ML skills, you are most likely well aware of the potential machine learning has to revolutionize various industries. But have you noticed that not all companies are jumping on the bandwagon? In fact, many companies, perhaps yours, are not doing much machine learning despite having some Data Scientists in their lines.

Some arguments I often hear are:

• “We don’t have the infrastructure for it.”
• “The quality of our data is insufficient.”

But I would argue that it’s almost never related to the organization’s maturity or incompetence, because both of these problems are straightforward to fix.

So why are some companies not adopting machine learning? The answer is simple: how companies make money is what drives their behavior.

Big Tech companies and machine learning startups do lots of machine learning because it impacts their revenue. That’s not always the case in other industries.

If your company is not making money directly from machine learning, it’s unlikely they will invest in it significantly regardless of their original intentions. Even if they are aware of the potential benefits, the costs associated with implementing and maintaining machine learning systems may not justify the return on investment at first sight.

What can you do about it?

If you’re an ML practitioner currently working at a company that’s not doing enough machine learning (or not at all), I would completely understand if you are feeling frustrated and asking yourself whether you should look for a new employer.

Although in principle neither option is wrong (staying or leaving), I would strongly suggest that you don’t rush this decision. By analyzing companies under this lens just described above, it’s possible to conclude that if you blindly jump to another job, chances are you’ll end up in the same situation.

So what can you do about it? Here are some tips:

Consider the opportunities

If your company is not doing machine learning, that doesn’t necessarily mean there are no opportunities for you. It’s important to evaluate the potential for career development and decide whether the opportunities available are beneficial for your career goals long-term.

Personally, working as a Data Scientist at companies whose business model was based on software is where I improved my software/data engineering skills the most, which have been incredibly useful. On the other hand, working at companies with other “non-tech” business models was when I gained tons of experience managing projects, budgets, stakeholders, and processes.

Analyze companies under this lens

Perhaps, after considering all opportunities currently available, you still decide the best you can do for your future is to look for a new challenge. In this case, understanding the business model of your potential employers can be incredibly useful for anticipating the kind of role you’ll actually have, regardless of the job description. You can also use this point of view to formulate questions while interviewing for new jobs.

Final thoughts

In this article, I’ve used this prism (how does company X make money?) to analyze why some companies are not where they said they want to be regarding ML development and implementation. But in fact, you can use the same lens to understand many other situations, such as:

• Which departments are likely to experience the biggest growth over time.
• Where you, as a Data professional, can have the largest impact with your work.

By applying this mindset, you can make more informed career decisions and set yourself up for success in the long run.