Enjoying your work feels like a life goal for many. For Yale Waller, data wiz at SingleStone, he loves his work because he intentionally chose his field. Find out why he switched university majors to pursue a career in data and the four pieces of advice he has for those looking to work in data analytics.
Meet Yale Waller, Data Analyst
Can you tell us how you got your start in tech?
As a finance major at William and Mary, I was set on attending law school after completing my undergraduate degree. During my junior year, I took an Intro to Business Analytics course and instantly fell in love with data.
More than midway through my degree, I switched gears and changed all my classes to support a career in data analytics. I haven’t looked back.
That’s certainly a shift – what about data excited you?
I’ve always been into technology, especially the emergence of artificial intelligence. When I realized that I really enjoyed working with data and that AI is within the same field, it was a no brainer.
It also occurred to me that data is increasingly running the world, so contributing to that space, helping make the world a more efficient place, was something I wanted to be a part of.
What are some of the positives of data “running the world”?
This might sound a bit odd, but I think one of the positives is taking decision-making away from humans. It’s like you’re pulling out the “oh, well my intuition says this,” and instead codifying it into a set of rules and processes that you can interrogate and improve upon.
Data helps you iterate and make progress faster. It sheds light on things you might not have noticed in the first place and enables you to ask questions that you otherwise couldn’t have asked.
And the other side of the coin, what are the negatives of data analytics?
Data itself can be biased and misused. People can use technologies that they don’t fully understand to make decisions. Certain types of models have assumptions baked into them. If you don’t meet those assumptions, you should not be using it and you can’t trust the results. But people use them anyways. It’s kind of like handing car keys to someone who’s never driven a car.
How can data practitioners train themselves to see through the bias?
Seeing bias in data is a learned skill. We try to make data practice a science, but a lot of it is an art. Seeing bias in data is definitely a learned skill and part of the art because you have to look in the negative space of the data and see what’s not there, more often than what actually is there.
What project is challenging you right now?
We’re helping a client leapfrog their data infrastructure. Their data is currently stored like the green text on machines that you see in James Bond movies, and we’re helping them move that into the cloud. All of this being done in a very aggressive timeframe. We’ve had to get a lot done and we’ve had to be very rigorous about it.
My background lies more in the machine learning side of things, and this project demands more data engineering skills, like working with databases and setting them up and end detail. I’ve really had to push myself and delve into the specifics of data engineering.
What’s your advice for someone considering a role in data analytics?
You need to do a lot of hands-on work. There’s an infinite number of PowerPoint presentations and infographics on data and machine learning, but you really need to grab datasets—the uglier the better—and make them look nice, clean them up, and then run models on them. Check out Kaggle, where you can access tons of datasets to test and grow your skillset. Really play with the data, move it between different stages so that you can see how all the parts interact. Learn how to understand and differentiate a data lake from a data warehouse and why you need different databases and what’s different about them. You really need to dive into the practicalities of it all.
I also recommend reading as much as you can. I frequent KDnuggets, for in-depth, technically useful articles. Sometimes I go a bit deeper and read research papers from arcserve and try to apply the research myself. At a certain point, you’re having to teach yourself these things, and if you’re trying to stay up to date on the field, you have to go to the papers and wait for a package to get deployed out there. Another thing that helps is browsing peoples’ GitHubs. People have done all kinds of really great open-source work with machine learning. Sometimes I’ll go to GitHub and type in neural network and see what people are working on – like handwriting recognition, fascinating stuff.
And then going back to biases. It’s important to learn the assumptions and requirements for different models or data sets so you don’t make those assumptions.
My final piece of advice is to get good at explaining complicated things to laymen who only have 30 or 40 seconds to listen to you. You want to be able to take this big, highly complicated thing and strip it down to what they need to know. The math is interesting, but the majority doesn’t care, they really want to know that it works.
What is your favorite part about working at SingleStone?
The people. Everyone is so friendly and willing to help you with whatever you need. There’s this culture of expertise, respect, and shared learning. The breadth of technical expertise and people you know you can rely on is impressive. One day I can talk to Brian Lipp about some data engineering thing that he’s uncovered and wants to give a presentation on. And then the next to be talking to our CTO Ryan Shriver about skill matrices for the organization. It’s really cool to work with people who are this down to earth and have so much technical expertise.
Feeling inspired to finally take the plunge and work in a field you know you’re truly passionate about? Does it happen to be in tech? We have open positions right now and would love to hear from you. Send us a message.