When Harvard Business Review came out with its article labelling Data Science the sexiest job of the 21st century, it grabbed a lot of eyeballs. Safe to say that that ~3000-word article initiates a mad frenzy as people scrambled to learn Data Science with the eventual hope of becoming a Data Scientist.
About 10 years later, that hype doesn’t seem to be dying anytime soon (if any, it has exploded exponentially).
But before you think about transitioning into this — admittedly — very lucrative field, there are a few truths that no one seems to be talking about, and which you absolutely must know before the eventual career shift.
A few years ago, if you knew at least a little bit about Machine Learning and Data Visualization techniques, you’d be hired for $100k+ jobs and get to work on interesting projects.
However, this landscape looks very different now for Data Scientists. Each year, universities, online courses and (dozens) of YouTube videos pump out tens of thousands of newly-minted Data Scientists. Besides them, many Data Analysts switched their LinkedIn “profession” to Data Scientist.
And all of a sudden, you see this huge bubble of Data Scientists whose count far exceeds the number of available jobs. As you can imagine:
Supply >> Demand
The title says it all. Some recruiters for Data Scientists jobs are terrible at spotting good talent: “Oh, you’ve got 6–10 years of experience? Let’s hire ya”.
Coming back to the first point: most small-scale companies literally aren’t ready to take on a Data Scientist. So whether their “data journey” is at the point of deploying Machine Learning algorithms in production or not, they must absolutely need Data Scientists because “Data Science is cool now”.
So while you might spend years perfecting all these Machine Learning algorithms and skills, all you might end up doing is manually filling in Excel spreadsheets (after all, a Data Scientist does work with “data”, no?).
It’s a clear example of Expectation vs Reality.
Every company is different, so I can’t really speak for all of them. But, many don’t have the necessary infrastructure in place to derive real value from data. This eventually lends itself to the cold start problem in Artificial Intelligence.
Oh, and if you see a prospective job specification that cites a long list (>5) of software and skills you must have, stay well clear: it reeks of a fundamental problem of a company that has no idea what their data strategy should be and their willingness to hire anyone because they think that “hiring any data person will fix all of their data problems”.
Emphasis on a lot. I suppose this point would make more sense to those toying with the idea of becoming a Data Scientist than it would to professional Data Scientists.
You’ll see a lot of multivariate calculus, linear algebra, statistics, probability, and information theory used in several Machine Learning algorithms and even in Data Mining.
So, if you’re terrified of math and disregard the sight of mathematical equations, you’re not going to have much fun as a data scientist. But luckily there are articles and courses that bridge this gap and make it a tad bit easier to transition into Data Science especially if you don’t have a math degree under your belt.