Q&A with Frank Pecjak, Senior Manager of Data Science with SCOT

Frank Pecjak grew up in Southern California but spent over 20 years in Northern Virginia after earning a BS in Information and Decision Systems from Carnegie Mellon University. After 8 years of analytics consulting, he went on to earn his MS in Statistics from George Mason University. He spent 14 years in cross-media and advertising audience measurement, building methodologies integrating panel and big data census data sources into a comprehensive behavioral platform. Frank and his family moved to Seattle in 2020 to join Amazon's Topline Forecasting team as a Data Science Manager.

What is your favorite Amazon memory?

It has to be my first year at Amazon. I started in the spring of 2020, right in the middle of COVID. We were updating our forecast every week and reevaluating almost daily, trying to keep up with the changing story and dynamics. Then, on top of that, my family was moving from Virginia to Seattle. Managing the move, saying goodbye to friends, and exploring a new city, school, and friends just made every day a new adventure. The team was so collaborative, smart, and welcoming. The team has remained that way throughout my time here, and I still remain in touch with many who have moved on to other teams.

How have you handled challenges and achieved success in the past?

I was fortunate enough to have mentors who reinforced early in my career that when in doubt, the data always tells the story. I've found that tough problems or challenges are best navigated with data points, starting from first principles and building from there. Obviously, that applies best in data analytics situations or ambiguous questions, but I've found it also works well in managing cross-team collaboration when you have data and hard examples to back up your take.

Can you tell us about your background and how you got started in the tech industry?

My bachelor's degree was in Information Systems, and my first job was doing data warehouse and BI consulting. I had the opportunity to work across various industries such as finance, publishing, CPG, and healthcare. That experience taught me how to learn quickly and establish patterns to find similar questions and approaches across seemingly diverse spaces. I eventually realized that what I loved most about the work I was doing was the analytics, and I wanted to expand my skills there. I got my master's degree in Statistics and transitioned to a methodology leader role in a digital audience measurement company. We transitioned over time to cross-media measurement, so I got a lot of experience looking across the tech landscape and understanding the dynamics of how audiences behave and the role of media.

Why did you choose Amazon?

I was first interested because it was a new domain and problem space to explore, but what really sold me on Amazon was the excitement and engagement I witnessed as I talked to people during my interviews. This was a culture that was proud of the science work they were doing and eager to push boundaries. That was a culture I wanted to be a part of.

What accomplishment at Amazon are you most proud of?

That's hard to answer because I never feel like we're finished with our work. Each iteration leads to new ideas and bigger plans. I am proud of the way our team has taken on the challenge of structuring planning inputs to our forecast. It's a tough problem because each team has a different way of thinking about customers and demand, and the metrics are different, but to have a scalable process, we need a rigorous model and process to translate into our forecast. The cross-team collaboration and the science development fit well with how I like to solve problems.

What brings you passion/excitement to your role and team?

There is nothing quite like holiday peak. From a forecasting standpoint, we spend a lot of time and energy in the months leading up coordinating with teams and running through scenarios and assumptions. To see it all come together and how our forecasts end up is amazing.? We’re also embedded in the real time daily rush of what is currently going on, why is it happening and what can teams do to push even more customer engagement. The cross-team collaboration and excitement to deliver is amazing.

Who has been the most inspiring influence in your life?

My dad was an electrical engineer in the aerospace industry for over 20 years, but he didn't even have a bachelor's degree. He had built domain knowledge and experience that made him uniquely qualified to work there, regardless of education. I always use that as an example and reminder that theory and education are important, but you get a lot out of hands-on work and iterative experience.

What do you do for fun/ work life balance?

I love watching sports. It really doesn't matter the sport; I can pick up and watch anything. My favorite pro sports are soccer (EPL), hockey, baseball, and NFL. My live watching favorites are anything my kids play. I'll become a fan of all levels of those sports, so now I'm obsessing over college lacrosse and volleyball rankings and keeping up with pro lacrosse.

What is your favorite leadership principle and why?

Are Right A Lot because at the core its really about listening and learning. With experience, you build the high judgement to know how to interpret and react to the information you get, but it always still comes back to seeking perspectives and challenging your beliefs.

Best place you have ever traveled?

I was fortunate early in my career that I traveled a lot. I don't know that I can name a best place, but the best visits I've had have been when I was lucky enough to tour and experience a city with locals. You end up away from the normal tourist areas and really get to know the city. Those are always the stories I remember and retell over and over again.

What advice would you give to aspiring?data science professionals?who are starting out in the field and what qualities do you think are essential for success in this industry?

I've recently seen a trend for data scientists to focus too much on coding, modeling complexity, and technical skills as a measure of growth and success. But equally important is the understanding of the business domain and problem you are solving, and tying it to the underlying analytical and statistical foundation that best solves the problem. The classic Venn diagram describing a data scientist as a balance between Computer Science, Statistics, and Business domains remains true, and I would encourage anyone starting out to cultivate skills in all three domains.

To check out opportunities with Frank's team click here.