Data Science is a new field and it is very broad. To keep things simple, the heart of Data Science lies in the implementation of complex algorithms to data. This is done to discover some underling relationships between variables to predict a certain outcome. The primary goal is to enable “intelligent” machine driven decision making.
I started my first job in 2012 with a large management consulting firm. I spent six years identifying operational problems, trying to quantify them and recommending ways on how to fix them. This often involved using conceptual models and simple statistical analysis. The technical term for these types of methods is “Soft Operations Research”. A lot of work and effort went into applying these methods and there was a significant reliance on qualitative input from individuals. These methods are still relevant today but in my opinion are more costly in terms of time versus value added. I believe the value they are adding to industry is diminishing.
It is because of this perspective I found the work I was doing at the time not very rewarding. It was time for a career change.
It made sense that my next move should be to branch into “Hard Operations Research”. This falls under the Data Science family. Lucky for me the university of Pretoria’s Industrial Engineering Department offered this as part of their Masters program. Yey!
The Masters projects the university provided appeared to be very complex. You can access them here. Though they were in the Industrial Engineering domain, I did not feel that my undergraduate degree had prepared me for the challenge. The Department recommended that I enroll in an online introductory Data Science course.
After six months I had completed six courses provided by coursera in the Data Science Specialization. These courses are pretty much very introductory. However, I found them useful as I was able to understand in more detail what the field was about and assess how well I can pick up coding skills.
The result was:
I never managed to submit my Masters proposal to the university of Pretoria. I enjoyed what I was learning on coursera and wanted to know more.
I was able to follow conversations in Data Science discussions and forums and build a few models (though I could not interpret them well at the time). I had some idea of what I was getting myself into. I actually thought I had a good grasp of the field and started applying for jobs.
Making It Formal
My first round of job applications and interviews was not successful. My decision to enroll in the first Masters in Data Science program at the university of Cape Town was to simply make myself more attractive to employers. I didn’t necessarily think the degree will significantly contribute more to my knowledge. I thought I could easily just do more online courses. I was wrong.
It was not about how well the methods were taught by the university lectures. The value was in the evaluation method. Most online courses are very lenient when it comes to assessing your competency. I found myself writing four hour coding exams, being tested more on my understanding and interpretation of the results, models and data. Let’s be honest, no one studies for an online certificate, there is google to help you out of a tough situation. The university post graduate approach enforced teaching one good habits as far as applying Data Science goes.
Getting The Title
The foundation received from the coursework was solid. Solid enough for me to get multiple job offers without full completion of the degree. I chose the job with the closest proximity to my house. Cape Town traffic is horrific.
Doing a coursera specialization alone would not have landed me my job. Well, maybe doing multiple more courses and I would have been able to talk my way into one. However, there is a risk to this. If you are taking the online route it is essential to make sure that you are surrounded by good mentors and solid leaders in Data Science. Same applies if this will be your first job after university with no post graduate specialization.
I encourage getting a sense of what the field is about before jumping onto the band wagon. You also need to know whether you have coding anxiety or not. Be prepared to do Mathematics etc. If your base degree did not have a strong math foundation there is a lot of work ahead. Unless you just want the title for glamour purposes you can ignore my advice. After all, this is not a regulated profession. The only requirement to being a Data Scientist is calling yourself one. Please do not be that person.
Becoming a Data Scientist was a good career decision for me. It is not because I love my job (a job is a job at the end of the day), more so that I have a skill that has no limits.
It does not do to dwell on dreams and forget to liveAlbus Dumbledore