Since data scientist is a senior job title that comes with significant experience, even expertise, how can you become a data scientist fresh out of college, or from Coursera classes, or from some data science boot camp?
I believe you indeed learn data science on the job. Employers have been fooled into hiring people with R, Python, SQL, NoSQL, some outdated statistical knowledge, and little work on a small academic data project - who claim to be data scientists. There's even a joke about NoSQL: some candidates add NoSQL to their resume to tell potential employers that indeed, they don't even have SQL in their skill set. The results is a painful learning curve for the new hire once on the job, and essentially negative ROI on the high salary commanded by these individuals.
Source for picture: click here
But employers could hire someone with the right background and turn him/her into a data scientist, a much better solution in my opinion. Someone with a quantitative background - maybe even a biologist - with experience working with various data sets in a professional setting, could be a better fit. After all, our own intern at DSC came with a nuclear physics background fresh out of postdoc, worked on great data science projects for us, and still helps maintain one of our core systems today, even after finding a full time job with another company.
It is true that data scientists should know some stats, machine learning and algorithms, R, Python, SQL, unstructured big data, elements of real time architecture and APIs, and distributed architecture. And be familiar with some vendors such as Tableau. But these skills can sometimes be found in geographers, physicists, biologists (think about bioinformatics), epidemiologists, even some MBA's. And self-learners can catch up quickly, especially if they already know a programming language such as C, C++ or Java. But focusing only on people who call themselves data scientists is a mistake.
Think about this: before colleges offered the first MBA curriculum, there were no MBA guys. Yet there were plenty of people doing the work of an MBA. Some of them became the first ones to teach in MBA programs. And even today, plenty of people do MBA jobs (like COO, CEO, business analyst) and don't have an MBA degree. And they sometimes do it better than people with an MBA degree. The same can be said about data science.
There are of course many flavors of data scientists, and those really producing innovative science and techniques (especially in automating data processes - from defining, finding, collecting, aggregating, structuring, cleaning, summarizing, refining data, to value extraction and operationalizing the decision process), are expected to have a different background than most business or industrial data scientists. But they represent a tiny minority; they are the unicorns that erroneously too many companies are chasing.
DSC Resources
Additional Reading
Follow us on Twitter: @DataScienceCtrl | @AnalyticBridge
Comment
Great post - particularly liked the MBA analogy
One can learn data science on the job. Our data science team at my job are all fresh out of University with PhDs and the field varies from computer science, mathematics, physics. None of their CVs mentioned data-science but their field of study is the domain where data-science all started even before the term data-science emerged as a popular term these days. They were hired on the understanding that they have the ability to learn on the job very fast & also understand complex topics which are new to them in fairly quickly & that's exactly what we have observed. They have become very proficient in machine learning pretty quickly, most had no prior knowledge of the domain however they've been exposed to the same fundamental math used in machine learning as the math they learnt in their post-grad courses.
© 2021 TechTarget, Inc.
Powered by
Badges | Report an Issue | Privacy Policy | Terms of Service
Most Popular Content on DSC
To not miss this type of content in the future, subscribe to our newsletter.
Other popular resources
Archives: 2008-2014 | 2015-2016 | 2017-2019 | Book 1 | Book 2 | More
Most popular articles
You need to be a member of Data Science Central to add comments!
Join Data Science Central