Staff Highlight: Zack DeSario

Zack DeSario joined CUNY Tech Prep in 2020 as our Data Science Instructor. Previously, he worked as a Manager of Analytics of a Presidential campaign, a Data Scientist at Google, and a professional poker player. Read more below!

Tell me a little bit about your career path.

I grew up lifeguarding in East Hampton, then became a Lieutenant Lifeguard. *laughs* I got my degree in Psychology at College of Charleston, South Carolina. I did graphic design for a little bit and eventually found myself playing poker professionally for seven years.??

How did you get into playing poker professionally?

Funnily enough, my best friend dropped out of Harvard senior year and started to do it professionally which inspired me. I approached it very methodically. I am going to do ‘xyz’ to make money - these are my specific parameters and this is how I am going to study. I had specific tools to help, poker trackers, etc. It’s important to mention that everything was data-driven. I did it for a really long time. I eventually got really bored and taught myself how to code in the meantime.?

What happened next?

I eventually moved to San Francisco and spent a year building on myself professionally. I did a Data Science bootcamp with Galvanize and they eventually asked me to teach at the bootcamp, where I taught for a year and half. Then, I worked as a Data Scientist at Google with a lot of projects related to Machine Learning and Natural Language Processing - the role was originally supposed to be for someone with a PhD.??

Then I got cancer. It was out of left-field and it was a total wake-up call. I realized I wanted to do things for other people. I left Google and started working at Need To Impeach as Director of Analytics which eventually transitioned to working as Director of Analytics for Tom Steyer, a candidate for the Presidential Election 2020. If he had won, I would have had an office in the White House, which would have been hilarious. That job ended a week before COVID hit. Now, I work for CUNY Tech Prep as a Data Science Instructor. I’m back in teaching and I love it.

What do you like about teaching?

I really like being able to spread how awesome and powerful data science is. It is one of the most innovative fields in technology. Data science is going to change the world - it already is!? It’s the new frontier and being able to share that, get kids stoked on that, is awesome.?

Back in the day, the most powerful tool was a sword or hammer. Nowadays, it’s a computer. Being able to know it and customize it is a huge deal.?

What does your day-to-day look like?

I get up at 6 - 7 AM and check the waves. Even in the winter, I’ll go in the water - without a wetsuit. Then breakfast and coffee. Then I’ll start the day by watching a lot of YouTube videos about data science, and reading Medium blogs about data science. I always keep an eye as to what’s going on with the data science world. 3blue1brown , Statquest , are good channels to check out.

Then, I will make tweaks to my curriculum and attend meetings. Then, I teach class. And go to bed early.??

Where do you think the tech industry is headed? What trends are you noticing?

There’s a lot more automation happening and pre-date machine learning algorithms. GPT3 is a really useful model to get familiar with. You can use that as a base and then build and train it with your own small datasets. It’s already really smart and you can make it specific to your needs. Folks should know that there are a lot of services that are coming out to help you with data science.

The tech industry is also headed towards the cloud. If I were to pick up a new skill set, it would be developing with the cloud. Being comfortable with developing in the cloud and getting used to navigating and setting up Amazon Web Services (AWS) and Google Cloud Platforms (GCP) is important.?

Another trend is the development of neural networks. They were invented in the ‘60s as a hypothetical - and now computing power is unlimited and they’re taking over the world.

What is a neural network?

I would describe a neural network as a version of the brain. They mathematically work the same - sending signals across to an axon and neuron, firing it off will set off a chain of events and process information. Neural networks are really good at natural language processing and image recognition - things that humans are also good at. They are a blackbox - we don’t understand WHY they work so well, but they do.?

What do you like the best about your work? About CTP?

I wish I had CTP when I was in college. It’s the best class a college student could ever take. We teach you how to get a job - and not just web development and data science, but practicing interviews and resumes. For me, having the freedom to teach what is most important is great - I am teaching topics that are useful and applicable in the real world. They don’t teach you how to get a job in college. The only other useful course that I wish I would have taken is how to do my taxes.?

What mistakes do you notice a lot of students making?

A lot of students don’t leverage the machine learning algorithms to get insights from their data. For example, if you build a random forest with columns of data, what does that show you? The next step would be to ask the model what is the most important column of data which can provide serious insight.?

Advice for students that want to get into the field??

Apply to CTP or a bootcamp course. Don’t be afraid to mess up and fail a lot. A lot of times I will build something and it won’t work. Get good at Googling how to do stuff - you don’t need a traditional degree to be successful in this field. I did it without one.

Also, you don’t need to know the inner workings of the ML algorithm to be a successful data scientist. You don’t need to learn how the tool works, you need to know how to use it. I teach how to use the hammer, not to build the hammer.?

Thank you so much!

This interview has been edited for length and clarity.