The Shrinking Window of Opportunity for Data Scientists
It may not be as easy to get the title today, but it’s certainly a lot easier than it will be in five years, guaranteed!
I’ve been doing data science for over a decade but it wasn’t until 2014 that I was officially ordained “Data Scientist”. At the time, I was working for a large hospital where doling out such cutting-edge titles didn’t come easy.
The grip on titles in the medical world is firm and figuring out where this new style of work would fit in the general food chain, at what level, at what pay, with or without profit-sharing, etc., takes time. Everything in healthcare moves slowly and that’s a good thing when dealing with life and death issues, but frustrating when you want your new employee badge to read “Data Scientist” (I know, vain).
After months of efforts from my department, my colleagues, and myself, we finally succeeded. I shared the news with a pharmacist colleague while standing in the cafeteria line and he burst out in laughter. He wasn’t being mean, he just thought I was making this all up.
Today, things have changed; the title has gained respect, visibility, along with corresponding financial perks. And that is why the window is closing…
Stand Out from the Crowd
The barrier to entry back then was low. I didn’t have a statistics or math degree yet and most of my knowledge was tricks learned from Kaggle competitions. I was also Epic certified so I knew my way around the 5,000 obscurely-named tables in our large EMR reporting database. Oh, I was also the face behind all the models I delivered, always listening to end-users and always there to answer any questions about what was going on behind the curtain.
That’s it, that’s all it took to stand out - some Kaggle tricks, a few Epic certifications, and compassion.
The barrier was low indeed but few saw this as the incredible opportunity it was. Those interested in modeling assumed all data sets came in neat, comma-delimited files, while those that knew the data, never heard of Kaggle or machine learning, and even less would reach out proactively to field workers and ask how they could help. I made my share of mistakes. People looked up to me and assumed I was always working with statistically representative samples and all my notebooks were reproducible and transparent, but that wasn’t always the case, especially at first.
The point is that it may not be as easy to get that title today, but it’s certainly a lot easier than it will be in five years, guaranteed!
Becoming a Data Scientist Today
If you’re on the fence about this career, now is a perfect time. The resources available today are so much better than five years ago and the importance of data science isn’t in question anymore. According to the United States Bureau of Labor Statistics, the employment of all computer and information research scientists is expected to rise 19 percent by the year 2026.
You can still get in without a specialized degree like statistics or bio-stats, and not all of today's jobs require masters, but eventually, they all will. So join Kaggle and start competing, and check out my favorite 6 data science classes of all time (https://www.viralml.com/blog-content.html?l=My-Six-Favorite-Free-Data-Science-Classes.html ).
If you are already a data scientist or want to really stand out, check out The ViralML School and the Applied Machine Learning Tract.
I designed these classes expressly to teach you how to look at the big picture, from the idea, the model, to production implementation and a universal UI so the entire world, not just those on GitHub will get to enjoy your work. Stand out today!