Data science and Fashion

Who would have thought that science and fashion could get along so well? Data Scientist, Sandra Greiss, after a period working at fashion startupLyst, currently works at Asos. In this interview she tells us about the wonders of her job and when she dreamed of becoming an astronaut.

Guest Post by Viviana Attard

 

Tell us about what you do.

Sandra : I am a Data Scientist. What we do mostly is building models. Depending on what you focus on, you can do a lot of things with data science.  You can work on recommendations, for example.  You can build models that recommend products to the users on Lyst’s website. From the website, you can see what they [the consumers] click on, where do they come from, and their purchases’ history, for example. That’s one branch of data science.  That’s not what I actually do. What I personally work on more is ‘classification’ which means that given some product, I write models that predict what [these products] are made of. “Materials” is one of the projects I have been working on. Once we define what a product is made of, we put a tag on it to classify it. For example, let’s say we know that this is a bag, and then we know that is blue and [that] it is leather, we want to gather all the other information on our product so we can make it easier for our customers to find what they are looking for.

 

You mentioned before that Data Science is varied and there are a lot of things you can do with it…

Yeah, classification is one branch. You can classify or detect colors, materials, categories with data science and machine learning. “Recommendations” is another thing that you can do. We built the new search on Lyst using machine learning. This is based on previous searches that were done on Lyst. It has been built by monitoring what our users clicked on the website. With machine learning, you basically train a model – which means that you teach your computer – to guess what people want or look for based on their previous behaviours’ history on our website. Basically, we try to replicate the same activity that happens in our brain.

You try to gather as much detail as possible and you train your computer to predict these things: to classify the products, to predict on what they [the users] are gonna click on, or the models try to predict what the [customer] like based on what they write on the search bar. With machine learning, you train the computer to understand all of these patterns and then you apply all that on the website.

 

How did you become a data scientist? Is this a path you always wanted to follow?

I actually don’t have any computer science degree. I woke up one day when I was 16 and I said to my mum: “I want to be an astronaut” and then my mum was like “No, this is not going to happen”. So I said, “OK. I will be an astronomer instead”. I studied Physics, Astronomy, and Astrophysics. I did Physics in Paris for my undergraduate, and then I did Astrophysics here in England, and then my masters and my Ph.D.  That’s when I learned how to code. Mostly, I coded a lot during my masters and Ph.D. When I finished my Ph.D., I realised that I didn’t want to stay researching in Academia. I tried to find out what I could do with my skills, and that’s when I discovered data science. It’s literally just one year and a half ago that I discovered what data science was, and that you could do cool things with coding, statistics, maths and logic skills.

 

And then, you ended up in fashion…

My obsession with shoes obviously helped with that. Loving fashion does make a difference when it comes to code for a company like Lyst!

 

What do you like the most about being a fashion data scientist?

That you can combine something that is quite creative and, in a way, a bit artistic with science, which is square and logical. I think that having those together (fashion & data science) is not something that you can find everywhere. I could have used my skills in finance, but I didn’t find it interesting enough. There was something that it wasn’t appealing to me while fashion definitely was. Especially if you have an interest in it, or you follow it. It’s also a combination I thought it never existed. I didn’t think that tech and science could meet fashion. Until I found Lyst, and I thought: ‘Oh, this is interesting!’.

 

Is there any specific aspect that makes you enjoy your profession?

What makes this job exciting is that [data science] is quite a fast moving industry. You learn so much all the time. It’s not that you are set to A and B for the rest of your life. Every few months there is a new technology, there is always something that comes out, there is always something new that you can use or [that] you can improve, a new way to make it faster, a new way to make your models more accurate. I personally find all these aspects really exciting. I guess being able to apply [these aspects] to another fast moving industry like fashion is always challenging and I find that great. I don’t like when something is just the same all the time.

 

It’s often said that technical roles are male dominated. What has been your experience so far?

I agree the data science is predominantly a male dominated field. However, I have been in the data science field for 18 months now and I have already seen a bit of a shift. More women are getting into coding and in the tech world including data science, which is great. If I ever get asked to go somewhere to encourage younger girls to come in this whole business, I’ll definitely go for it because I think people need to understand there is nothing that females can’t do. You know, people, in general, do have stereotypes on how data scientists look like and the majority might fulfill these stereotypes, but there is still room for us to say that actually, it’s not completely true.

 

Why do you think there is still a gap between women and men employed in this field?

A lot of the men – not all of them – you work with can be very intimidating. Obviously, things now are changing. In the past, it was in this way. When it happens, they basically question your decision of being a coder, a data scientist or scientist. I think it’s a shame. Luckily, the female data scientists I know are basically at the same level as their male peers.

 

Where do you see yourself in the nearest future?

I would like to progress as much as I can in the data science field, but I’d also love to have my own business one day. I don’t know what, I need an interesting idea, but I haven’t come up yet with anything that is a problem that needs a solution for it. Until this happens, I think I am just going to focus on becoming a good data scientist. I want to understand more until I became a lead data scientist.

 

Is there anything else you enjoy besides coding? 

I love baking, travelling and eating out. I have got this obsession of trying every kind of restaurant in London as well, which it doesn’t help. I love to try a lot of different things. Recently, I have been liking brunch for some reason so I am trying all the different brunch places that I can come across London. East London has a lot of them.

I went to this really nice place by the canal in Haggerston. It’s called the Bargehouse on the canal. They serve the food in a nice sourdough that they fill with the filling that you chose, their cocktails are really nice as well.

 

Sandra Greiss
Photo by Liz Gregg

 

 

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s