The data career entry has become extremely difficult as compared to when I started in 2010. Back then anyone with any knowledge of data was welcome. There were people with extremely mediocre skills who were being promoted to data managers from data administrators. We never knew the power of what we were being exposed to. As a black African, I thought I was being pulled down by my non-black African superiors. It is now that I look back and realised that there are now barriers to entry. I feel for anyone who is passionate, has the skill, and is constantly overlooked.
In our last article, part 1, I shared four qualities that could assist anyone from being overlooked. These qualities or capabilities included (1) having mathematical fundamentals, (2) learning how to program a computer, (3) doing data wrangling and visualisation, and (4) being comfortable with statistics and probability. This article expands on this by giving you more qualities that could be game-changers nowadays. The beauty of it is that one can get all this online thus making a strong case whether one still needs a university qualification for a data career. That is an argument for another day. Let’s rather discuss others here.
Understanding Databases
My understanding of databases started in 2006. I used Microsoft Access when I had an assignment for the business information systems (BIS) module. This module was needed as part of my BCom finance qualification. Attending university allowed me the opportunity to learn complementary skills to finance. Back then I saw no point to lean databases. It was when I started working that I saw the power of Microsoft SQL Server Management. I had a support group to assist me with my learning. Storing data in databases is better than using text files. This is the reason why most companies are using relational databases instead files. Learning this will add value to the company and thus your employment. I would learn about converting files to databases if I was to start all over again.
Learn Machine Learning
My first encounter with machine learning (ML) was daunting. I enrolled for the optimisation elective module for my post-graduate in Informatics. I walked in from work, all formal, into a class full of people wearing sweaters. Only to encounter a screen that seemed like a Nokia snake game. I was told that was my assignment for the semester. I had to successfully program something that will learn not to crush on itself. There are lots of terminologies being used by this gang of people. These terminologies are important for effective communication. They speak of supervised learning, unsupervised learning, time series, or dimensionality reduction (clustering) techniques. Some are close to big data thus learning Hadoop and ML will be beneficial. This is the same reason why I chose both Big Data and Optimisation elective modules.
Conclusion
I was terrified when I was doing my honours (post-graduate) degree in Informatics. I sometimes wanted to drop off both Big Data and Optimisation modules because getting a passing mark was more important that learning difficult concepts. I succeeded with Optimisation but now it’s difficult to learn ML unsupervised. Find your own rhythm and get help to learn as much as possible. Irrespective of how much you want to quit and rather watch YouTube videos about money and wealth. It is important to find a tribe of friends who wants to achieve a career in Data Analytics/Science. This way you will be helping each other.
