Skip to content

Becoming a Data Scientist

2018 August 8
by Shannon Lowder

In my latest professional change I’ve moved into a role with a bright and shiny title: Data Scientist. I’ve heard all the hype around Data Science, Machine Learning and Artificial Intelligence, and I don’t buy a lot of it. I do believe these words will change how we do everything. I just don’t know that there is a perfect way to get from where you are to where the hype train is currently. I want to share my journey with anyone who wants to read it. Hopefully, you’ll learn something that will save you time and energy in your own journey.

Growing up, I had a knack for mathematics. I wasn’t the best student. I pretty quickly would learn what was being measured and just do those things that would get my grade to where it needed to be. That usually meant no homework, but plenty of projects and tests. By the end of High school I’d taken all the math courses available. In college, I continued taking math courses, but ended shy of a Math minor. I worked my way through school, so I made the call to finish out my Computer science degree, and forgo the last semester that would have added that minor. I did pick up two statistics courses. To be honest I hated them, since many of the examples used always seemed contrived and unrealistic.

One of my first roles I worked while in college involved building out the database for the backend of a book reseller. I built the database and access layer in VB6. Then others on the team built all the rest. But that first real database role would set much of the course of my career for almost a decade, I was just a programmer who happened to work well with relational database systems. It wasn’t until 2010 when I attended my first SQL Saturday event.

I attended this event where people were freely teaching others to do exactly what they did. They also encouraged each and every one of them to step up and share something they had learned. This was where I first met Geoff Hiten, Andy Warren, Andy Leonard, and many more MVPs. That very next week I took as much as I could back to work and applied it. The results were amazing! I was hooked. Before the end of the year I hit Raleigh and Columbia, SC’s events. By the end of the Columbia event, Andy Warren pulled me aside. He wasn’t asking me to give back and share…He pretty much told me to share something I knew.

That began my blogging and presenting career.  The thing about blogging and presenting.  It drives you to learn more about the materials you’re presenting.  So the more I blogged and spoke, the more I studied.  The more I learned, the more I wanted to learn.

This led me to get deep into Database Administration, and then into Business Intelligence.  In that transition I started applying the art of automation to BI solutions.  I would visit Andy Leonard’s blogs and classes as often as I could and by late 2011, he tried to convince me to learn this new language Biml. In learning Biml, you start thinking more about the patterns in the solution than you do the solution itself.  It was this change that I finally understood Data Science.

There are patterns in data.  One I saw that all ETL processes follow patterns that it dawned on me. Patterns can be expressed in functions.  Functions can be expressed in mathematical formulae… the processes we’ve been building by hand all these years could ultimately be automated with a narrow artificial intelligence.

What do I want you to learn from this?

If you want to get into Data Science, first you’re going to have to love math.  If you loathe math, and you can’t force yourself past that.  Data Science isn’t for you.  Primarily you’ll work with statistics, but I’m seeing a lot of discrete math too.  Discrete math is all about patterns, and modeling real world problems in terms of mathematical terms.  If you haven’t had any experience, or if you want a refresher with either of these, edX.org has some courses i’d reccomend before going back to a formal school.  That way, you can get a taste of the material before laying out a lot of money.

Secondly, I would highly recommend getting connected with a professional group affiliated with PASS. They have virtual chapters as well as in-person chapters where you’ll get connected with the most welcoming professional group you’ll ever find. These people will share any knowledge they have with you. Some of them will even partner up with you and mentor you. This group will help you get to where you want to be, even before you’re fully aware of where your destination lies.

The final area i’d start researching is patterns. Of course I’ll point you toward the patterns that caught my attention first: Andy Leonard’s SSIS Patterns.. But that’s no good to you if you haven’t spent the last few years building SSIS packages. There are patterns in every industry. There are sales cycles in retail. There are phases to the growing season in agriculture. Learn to spot these patterns. It’s very important to Data Science. I’m still looking for good material to teach this skill.

In my next article I will begin to share my experience on my first official Data Science project: Predicting the likelihood a given property will experience hail damage in the upcoming year.

No comments yet

Leave a Reply

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS