Sep 15, 2026 · 20 min · 11 segments
Ask most people what they think it's like to be a data scientist and you'll get the same picture: equations, spreadsheets, someone quietly crunching numbers in a room by themselves. Nadya DeBeers has…
So you're working in a real estate startup and you said part of what you do is you look at predicting how quickly something will sell.
If you were to break that down into sort of small components about how you do that for people that don't understand what it is.
What is that? So I would start with the problem statement, which is how can we predict if, let's say, the house next door might sell in the next 90 days? So as a human on the streets, I know my neighbor.
What are the signs I would look for? Maybe they start tidying their front yard, right? And they've never done that before.
But what I do is take the information I do have and try to translate it into signals that a model can then learn from.
So a human can do modeling in a sense, but a model can do that work much faster and it can identify patterns that we might not pick up.
You're the one, as the data scientist, actually doing a lot of that manual thinking and upfront work, identifying what a signal might be, creating that signal from the data you have available, feeding it to the model, and then doing something with that output, which would be the
across many many industries so try to look at prediction whether it's real estate like what you're looking at and i know that they do a lot of prediction of the number of people that would be on certain train carriages as train is coming through and so
And that is something that drew me into data science because like I mentioned, I have a lot of passions.
But the great thing about data science is if there's a website, if there's an online component, there could be a need for a data scientist.
So that means I could work in any field that exists, which really excites me.
So you're working in a real estate startup and you said part of what you do is you look at predicting how quickly something will sell.
If you were to break that down into sort of small components about how you do that for people that don't understand what it is.
What is that? So I would start with the problem statement, which is how can we predict if, let's say, the house next door might sell in the next 90 days? So as a human on the streets, I know my neighbor.
What are the signs I would look for? Maybe they start tidying their front yard, right? And they've never done that before.
But what I do is take the information I do have and try to translate it into signals that a model can then learn from.
So a human can do modeling in a sense, but a model can do that work much faster and it can identify patterns that we might not pick up.
You're the one, as the data scientist, actually doing a lot of that manual thinking and upfront work, identifying what a signal might be, creating that signal from the data you have available, feeding it to the model, and then doing something with that output, which would be the
across many many industries so try to look at prediction whether it's real estate like what you're looking at and i know that they do a lot of prediction of the number of people that would be on certain train carriages as train is coming through and so
And that is something that drew me into data science because like I mentioned, I have a lot of passions.
But the great thing about data science is if there's a website, if there's an online component, there could be a need for a data scientist.
So that means I could work in any field that exists, which really excites me.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.