Data Science at Syrup: Get to Know Lorena

Data science forms the foundation of the work we do here. Get to know one of our outstanding data scientists, Lorena Piedras, in this Q&A.

Where is your work-from-anywhere office?

I'm originally from Mexico City. I lived there kind of all of my life growing up. And then I moved to New York around two or three years ago for a Master's. And now I still live here. I graduated, then started working at Syrup and stayed in the city.

How did you get started in data science?

I studied actuarial science for undergrad, so very heavy on statistics, and I really enjoyed the probability part of it, linear algebra as well. And I had a couple of algorithm classes, and I liked that — those classes, I think, were my favorite. So it was just kind of the combination of things that I liked that fit in really well with data science…it was just everything that I enjoyed!

What brought you to Syrup?

I really wanted to go back to working in a startup. I find it very exciting to build something from scratch. And also the ownership and accountability you get there — that things work out because you're doing a good job…or not.

The impact that you can have is so much bigger, which is something that excites me and kind of scares me a little bit as well.

The other big part is I wanted a company where the function of data science, or what I did, was at the core of the product. And Syrup's Forecasting Engine is a really big part of the recommendations that we make!

And the last one was more team related. I remember in the interviews, it felt like a collaborative team solving session. It didn't feel like it was a panel where I was getting looked at and kind of interviewed…which I was. It just felt very natural, I liked working through the interview questions with the people that interviewed me. So yeah, that's the other part.

What's it like working with AI every day?

It's very exciting because there are a lot of changes happening — a lot of the recent impact and benefit that we've seen from AI right now has been from research that has existed for 30 years. But it's very exciting, exciting to be at a point in time where it's actually being used a lot and it has a lot of impact on the business. Just working on that feels very rewarding.

And on the other side, it can be quite intimidating because it means that things are moving so fast. It's hard to keep up with the pace of change. So it just means spending time researching, reading, understanding what's new. And that's part of the field changing so quickly.

I'd say exciting and intimidating in summary.

Tell us about an interesting project you've been working on.

I'm working on the hypothesis or question of, if we use a more specialized model for subsets of the data — it could be using one model by channel, using one for e-com and one for retail, using one model for high selling versus low selling items, or for items that are quite new versus others that are never out of supply — would that give us a boost in performance?

So it's thinking about the different subsets of data that make sense to have a model trained to them.

And then of course there’s the training and evaluation part of actually testing the hypothesis that we have on whether that brings us better results or not.

What do brands and retailers gain from having a team of data scientists supporting them?

So we look at their data a lot. And sometimes we notice certain anomalies in the data, things that don't look right. And that's helpful to them because they can correct or fix it before it becomes a problem.

The other part is, as I said, to build a model, we need to look a lot at the data. So we have a good intuition about sales trends and behavior of different stores, SKUs, and stuff like that. So we can also provide insights into that. I think it's super helpful.

And yeah, the last one is the team has seasoned people that have worked on forecasting for retail in specific for a lot of years. So there are many interesting ideas on external sources of data or additional problems that can be solved through data science that our customers can benefit from as well.

If you could swap home bases with anyone at Syrup, where would you go?

I guess the most common answer is Maui, but I don't love the beach. I kind of do, but it's just not my vibe. I like mountains more. I know Tori just introduced herself today and she lives in the Catskills, so I think that would definitely be a place that I would like living in. Just because I like hiking a lot, I like trail running. It would be lovely to have the mountains there and be able to do it in any given day. I'd really enjoy that.