Common Misconceptions of Working in Data Science w/ Sarah Nooravi (Episode 12) #DataTalk

Listen to the podcast:

Every week, we talk about important data and analytics topics with data science leaders from around the world on Facebook Live.  You can subscribe to the DataTalk podcast on Google PlayStitcherSoundCloud and Spotify.

In this week’s #DataTalk, we talked with Sarah Nooravi about what it means to work in data science and some common misconceptions.

This data science video series is part of Experian’s effort to help people understand how data-powered decisions can help organizations develop innovative solutions. To suggest future data science topics or guests, please contact Mike Delgado.

Here’s a complete transcript:

Mike Delgado: Hello, friends. Welcome to our weekly Data Talk. It’s a show featuring data science leaders from around the world. Today’s topic is common misconceptions of working in data science, and we are super excited to have Sarah Nooravi here with us. She went to UCLA, got her Master’s of Science degree in mechanical engineering.

Got her Bachelor’s of Science in both mathematics and economics, and a minor in statistics. Sarah, I don’t know how you did all that as an undergrad. And she also teaches at the DataViz Bootcamp at USC. She leads monthly machine learning meet-ups. She is super active and helpful on LinkedIn and I highly, highly recommend that you follow her. Sarah, thank you so much for being our guest in Data Talk today.

Sarah Nooravi: Thank you, Mike. Thank you so much for having me.

Mike Delgado: So Sarah, I just can’t believe all the undergraduate work you did. You know, both mathematics and economics, and a minor in statistics. How did you manage that workload?

Sarah Nooravi: So, I think maybe to clarify, it was a math econ degree.

Mike Delgado: Okay.

Sarah Nooravi: Yeah. And then a minor in statistics. It was a heavy load, but I think it was … It wasn’t too bad. I think I really enjoyed the classes that I was taking, and so really being able to kind-of enjoy it as much as kind-of engage in it was … yeah, pretty good.

Mike Delgado: And so after you finished your undergrad, you decided to pursue a master’s in mechanical engineering. Tell us about that process.

Sarah Nooravi: Yeah. So, I think this is an interesting kind-of carer choice or educational journey that I took. So after I graduated with my math econ degree and my minor in statistics, I knew that I would probably end up in either a consulting job or a data analysis job. And what’s funny is that I kind-of only considered one job out of college.

I really only applied to one place. And it’s kind-of … The reason why I only applied to one job is ’cause I have a strong passion. I think all throughout my time at UCLA, I realized how much I really love people and helping people. And so I think I volunteered a lot during my time as an undergrad, and really was looking for a job once I graduated where I could really help people. You know, how doctors can help people one person at a time?

Mike Delgado: Mm-hmm (affirmative).

Sarah Nooravi: Really was hoping to kind-of … Okay, so my … I’m clearly not in the healthcare field. How can I utilize my math skills and really help people around the world? And so I was really passionate about working for this nonprofit that worked with … kind-of partnered with NGOs in third-world countries. And basically to help them remove the gut-based decision making and kind-of go more towards more of a data-based policy making. So I was really interested in that. And unfortunately, it was all the way on the east coast, and my mom wasn’t too fond of my moving to the east coast.

So I was like, okay. Well, next step. What can I do, right? How do I … again, do I leverage my math skills to help people for the greater good? And really at that point I had gotten interested in nuclear fusion and renewable energy. And I was like, okay. Let me leverage my math skills and apply it towards engineering.
So, and this is any career change, I think. Most of the people on this Data Science Talk are kind-of going through transitions. And really, it takes … maybe we can talk about that a little more.

I think it’s very hard to go from not being a mechanical engineer to forcing my way into kind-of the program where I’m with a bunch of mechanical engineers. I’m the only econ person there. And so, I got interested in renewable energy. Did some research there. And then upon graduating, really at that point it was like, okay. Do I leverage my mechanical engineering degree, or do I leverage my math econ? And that kind-of brought me into the data science space, because the job market is what it is. And I was happy to kind-of pursue either direction. I think I was … Yeah. Very happy to kind-of take either direction.

And if I could take this one step further, I mean, really when I made the choice to get into data science, it was more along the lines of, where do I fit in culturally. Because when I interviewed at Operam, I saw guys in scooters and roaming around and like an open office, and it was a small team. And I was like, hey. I could be at home here. And so working on complex problems was something that I loved and they would provide. And then working with good people, people that I could see myself enjoying myself coming to work with, I think was very important. So …

Mike Delgado: That’s cool. So, if you were to go back and do college again, would you do anything differently? ‘Cause I think will be important for the data science community to hear from you on this.

Sarah Nooravi: Yeah. I think … Okay. I think my one regret is not emphasizing more on English.

Mike Delgado: Really?

Sarah Nooravi: Yeah. I know that sounds odd, but I think as a part of what we do, a lot of it is communication. And having more English under my belt, I feel like I probably would do a better job at maybe communicating, not only just verbally, but non-verbally, right, through … whether it’s documentation or whatever, or even through the posts that we do or through articles. Like, I feel like it really helps when you’re able to not only communicate verbally, or even in your own head, but like on paper. And so … So yeah.

That’s one thing that I’m maybe even hoping to go into, you know, and take some courses in English, just because I feel like I had focused so much on the math side of things, that being well rounded is super important in this field. So I feel like having exposure to a bunch of different fields could be very beneficial.

Mike Delgado: That’s funny you say that, because whenever I see your posts on LinkedIn, you’re so thoughtful. You’re very well written. So it’s funny you say that you struggle with that. Because whenever I read your content, I’m like, wow. You know, you’re able to communicate things so simply and even I can understand it. I don’t come from the data science field.

Sarah Nooravi: Yeah. It takes time. Like, could I do the … could I optimize how fast write my posts? But yeah. That’s, I think, something that I’ve kind-of developed over time. I don’t know if we touched on that, but kind-of the teaching that I did throughout my career journey and even .. yeah, just through my educational background, I did a lot of tutoring, and then eventually the teaching through Trilogy.

And I think it’s definitely a very good experience to learn how to communicate well, and communicate the same concept over and over to different people who understand things very differently. Right? So some people are visual. Others are, like, okay. Tell me the theorem and explain to me in a proof. And so understanding that people have different learning styles really helps, I think, in kind-of developing this communication to help people understand.

Mike Delgado: Yeah. I mean, especially I think about my learning style. I’m an auditory learner, and I’m also a visual learner. So I like to see visuals. So like, appropriate charts and graphs are helpful for me. Yeah. I mean, we all learn differently, right? So … And part of your role is trying to figure out, how do I tell this data story in the most effective way possible to this group of people.

Sarah Nooravi: Yeah. Yeah.

Mike Delgado: So, I would love to hear your definition of what is a data scientist before we move on in this interview. And I was funny. I read this tweet by Josh Willis who said … he kind-of had a joking tweet. He said, “A data scientist is a person who’s better at statistics than any software engineer, and better at software engineering than any statistician.” Pointing to the fact that data scientists is truly that unicorn. Right?

Sarah Nooravi: Yeah.

Mike Delgado: I would love to hear your kind-of… you know, having worked in the field, teaching data science, how do you define it to like your friends and family who are going, like, “Sarah, what do you do?”

Sarah Nooravi: Yeah. I think this is funny, ’cause like, I do have to explain it a lot to my family. They’re like, “Sarah, you just work with numbers, right?” I’m like, okay. So, I mean, I think it’s interesting, because when I do try and explain it to people, it’s usually from … it depends on who they are and what their background is, right? So if I’m talking to someone in healthcare, it’s like, how do I relate this back to, okay, x-rays, MRIs, like you know, cancer detection. Things that they would understand. If I’m talking to someone in marketing, it’s okay, how do we … it’s basically how we would optimize the marketing campaigns, right? So like …

But in the broadest sense, it’s … the way that I’ve, I think, kind-of succinctly would kind-of phrase this is that … or at least to me, is like, what I see is like, okay. We’re trying to remove gut-based decision making and replace it with data-based decision making.

So like, at every stage … So even with the nonprofit that we were working with at Operam, even at Mobility where it’s like, how are we removing kind-of this, “Oh, I think that this is the strategy that we should take moving forward,” and kind-of giving the why behind why we should pursue that business strategy. It’s like, okay. Yeah, we think that this is what we should do, but why are we doing it? And the why comes from the data. Right?

Mike Delgado: Yeah. Do you … you know, as you’ve been working in the field, how often when you’re presenting the data and telling the story … Do you ever get pushback from people saying, well, I don’t think that’s right? Like, my gut tells me this is not right?

Sarah Nooravi: Yeah. So, that’s actually pretty funny. So you lead into one of the misconceptions that I was going to mention, which is that people actually want … are open to hearing what you have to say. And really, at the end of the day, it depends on who you’re talking to. And most of the time, or some of the times, you’ll run into people who are like, “Yeah, that’s great, but it doesn’t align with what I expected. And so I’m not going to” … you know.

And you will get push back on that, and that’s actually more common than we think. So I think the way around that is really building up that trust. Right? So they may be doubtful, number one, either because they’re stubborn and they only want to hear conclusions that align with what they thought. Or that they just don’t trust the data. Because in a lot of cases, the data could be pretty messy. And so maybe a conclusion if you look at it one way could be different than if you look at it in other ways. So they’re like, “Okay. Keep digging.” You know? Like …

Mike Delgado: Yeah, yeah.

Sarah Nooravi: Like, okay. But, so I really think it’s a matter of taking them through the process of, okay. So, you know … and walking them through and … in order to build up that trust and have them not question you on everything.

Mike Delgado: I think that’s a really valuable skill, building trust. ‘Cause that’s really so crucial to get anything done in an organization. They have to trust you, trust your data analysis. So when you’re going into these meetings, are you kind-of already preparing yourself mentally that there might be this push back and how you might respond?

Sarah Nooravi: So, I haven’t had to present anything in awhile, but typically I don’t really expect pushback, because I feel … at least in my case, in the cases that I’ve had to deal with, I haven’t had too much pushback. It’s mostly I’m dealing with clients that are very … actually, I’ve been very lucky … who’ve been very open-minded. Right? And they don’t have access to the type of data analysis that I am providing.

And so they’re actually very happy. They’re like, oh, okay. This is really good information. But it’s always, can we take this one step further. Right? Like, okay. This is great. And so the challenge then becomes, okay. How do I collect the data and kind-of answer this problem in the way that they kind-of … and in order to provide the best kind-of … whether it aligns with what they were expecting or not, but just like a proper vetted approach to solving their problem.

Mike Delgado: So, you know, today’s show is all about misconceptions of working in data science. What would you say are some of the common ones you hear about regularly, things you’re constantly having to explain to friends, or even people in the workplace who think your job’s to do something else other than what you need to be doing?

Sarah Nooravi: Sure. So, I think the one that I get maybe a lot more on LinkedIn than anywhere else is that, hey. I’m coming from X background and I want to transition into data science. How do I do that? Do I have to have … like, and the key here is that it’s almost like it’s implied that the educational background that they have isn’t good … isn’t the one that they need, right? And os maybe it’s more implied, hey.

We need a computer science or a math degree in order to be a data scientist. And so, here kind-of what I want to highlight is that data science is not only just very broad, but the core of what data science is, is understanding how to take a complex problem, break it down, know how to deal with a large data sets, model, know how to code, use some creativity to attack that problem. And really at the end of the day … so the problem solving aspect of it is the biggest one, right?

So if you look at it, a lot of engineering backgrounds have taught you how to problem solve. Right? That’s basically what you do. Econometrics, neuroscience, atmospheric sciences, physics. Like, all these backgrounds have this fundamental core principal where you’re taking a problem and you have data. And you need to come up with some creative way of solving that problem. And then at the end of it, communicate it to someone else. Right?

So like, we don’t think that maybe, hey. Okay, maybe they haven’t been exposed to specific machine learning techniques or specific languages that we use in data science. But, as long as you have that problem solving mindset and the problem solving skills, you’re probably at a pretty good starting point to kind-of jump into just learning the tools now. It’s a matter of just learning to tools, versus the other way around, where maybe you don’t come from a STEM background and you’re trying to transition in. You’ll probably be exposed to the tools first and then have to learn along the way how to use them to problem solve.

So, yeah. You’ll find many data science teams, ones that I’ve worked on, where one was from atmospheric sciences, one was from physics, one was from business, others were from finance. And each person brings a very different lens of how they approached problems in the past that can be leveraged on a team, which I think is very good. One story that I have on this is that I was working on script analysis at Ope ram. And it was me and this Ph.D and we were working on the same problem from different lenses. And he was taking maybe the more mathematically rigorous approach. And I was like, okay. Well, how can I be a little bit more creative in my approach. And we had someone on our team who came from an English lit, creative writing background.

Mike Delgado: Really?

Sarah Nooravi: Yeah. And she was on our analytics team. And I was like, hey, girl. Let’s talk.

Mike Delgado: That’s awesome.

Sarah Nooravi: I’m like, what do you know about scripts? Let’s talk scripts. And just my discussion with her, I was like, okay. Perfect. I have a creative way of solving this problem now. And so, the domain knowledge that you have coming from different fields and transitioning into data science could really actually work to your benefit and give you a different lens into solving these problems.

Mike Delgado: That’s so cool. And I think what’s neat about that, Sarah, is that you’ve mentioned just a couple times the importance of creativity. And I think that’s also one of these misconceptions. At least something that I’ve had in that past before knowing much about data science was that, well, data science is mainly, you know, having stats backgrounds, working in programming, coding, and that’s what you do.

But what I’ve realized, and just from listening to you, creativity is a huge part, plays a huge role in the work that you’re doing.

Sarah Nooravi: Yeah. I think anytime that you’re approached with a problem, it’s not always directly obvious how you’re going to solve that problem. Right? And sometimes the obvious approach leads nowhere. And so you have to not only just understand the data that you’re dealing with, but once, and once you’ve understood it, kind-of formulate, okay. Like, we need to solve this problem. Like, how are we going to do that in a way where we’re leveraging all of the data but also being creative in our approach.

And so, yeah. I think it’s a little less known fact that creativity really plays a pretty large role in how effective you can be as a data scientist. Because you can do things. But the creativity, I think, is the differentiator between someone who’s just using the tools to achieve a goal, and then the ones that are using it effectively and coming up with very … like, and sustainable models and visualizations and stuff like that.

Mike Delgado: If you’re interviewing, ’cause we have a lot of people in our data science community that are looking to get their first role, their first job in data science, knowing how important creativity is, how would you interview or figure out how creative a certain person is?

Sarah Nooravi: I think for me, it’s popped up in kind-of how they’ve explained to me previous projects that they’ve worked on. Right? So if someone’s like, did some project, I mean, I would probably press them on, okay. Well, what approaches did you take? How did you think about the problem you were trying to solve and the data set that you were given? And kind-of walk me through your approach. And if there’s really not much there, then there’s not much there. But if they were like, “Yeah, and then I ran into this challenge, and then I ran into that challenge, and then this is how I overcame that,” right? Then you can start to learn where that creativity is coming from and if it exists.

Mike Delgado: That’s cool. What other misconceptions do you get or hear about?

Sarah Nooravi: So, the other one that I wanted to mention, maybe it still comes from this … from the more educational side. But that the misconception that you need to have a Ph.D in order to work in data science. And really, let’s just put it out there, most data scientists, I don’t have a number for you, but don’t have Ph.Ds. Most of them will come from like a master’s or a bachelor’s will be just fine. And I think really, this misconception comes from the fact that when you’re thinking about either getting a Ph.D or you have a Ph.D and you want to work in data sciences, where you see yourself, what you see yourself doing in the future, right? Is it developing algorithms from scratch, working for these big companies? Or is it doing more of the data analytics side of things? So if you look at the way that Facebook structured their marketing research team, they have their data engineers, they have a data science core, and then they have a data science analytics. And it’s really understanding the differences between what these different groups do.

And then, I mean, kind-of going to maybe smaller companies, mid-size companies, I mean, you’ll … I don’t think at any of the smaller or mid-size companies you’re going to be developing algorithms from scratch. You may be tweaking things. But it’s really understanding what you want to do in … as your end goal, and then aiming towards that. So if you’re looking to be in the data science analytics realm, then really a Ph.D isn’t going to be necessary. And actually, years of experience could play more in your favor than actually going and pursuing a Ph.D.

So, and something I’ll say here is that really evaluating someone’s learning skills are just as or even more important than your educational background. So if I see that someone can pick up on things very quickly, they’ll probably be a better candidate for working in this space than someone who’s kind-of stuck in their ways and isn’t willing or able to kind-of pick up on new things. Because at the end of the day, new libraries, new tools, new … like, new methods for solving things are coming out ever day. And it’s being able to pick up on those things kind-of use them in the business setting is very important.

And then there’s also this distinction between kind-of the Ph.D coming from an academic setting, right, and that transition from academia into industry. Where in academia you’re kind-of working at a slower pace, maybe you have years to develop something. Whereas in industry, they needed something yesterday. So it’s … Yeah. So it’s much more fast paced. So really, in this case, I would say, don’t worry about needing a Ph.D. Just focus on how well are you picking up on new tools and kind-of developing that. And then going and getting an internship. If it’s … if you’re just breaking in, really, an internship could work hugely in your favor and getting some experience under your belt.

Mike Delgado: I think one of the key points you’re mentioning here, Sarah, is just the importance of continuing to learn and adapt.

Sarah Nooravi: Yeah.

Mike Delgado: And one of those misconceptions could be, like, well, once I get my stats degree and I’m off. I don’t need to keep studying. And you’re pointing out, no. Like, it never stops.

Sarah Nooravi: Yep. Yep. That’s what’s challenging and exciting. It’s challenging because you’ll always feel like there’s something that you don’t know, and that’s just the matter … like, that you can’t change that. Right? There’s too much. We’re drowning in new technologies and new findings. And so you’ll never feel like you’re up to pace.

But it’s also exciting because there’s always something new to learn. And there’s always new libraries that you can be trying and exploring. So yeah. You have to have a very unique personality to kind-of cope with the challenges, but the excitement that come along with it.

Mike Delgado: Sara, how do you personally kind-of keep up with this, you know, huge learning curve that’s constantly happening with new libraries, new things happening?

Sarah Nooravi: So, I think for me, the way that I get around it, which is kind-of why I’ve started these meet-ups … So essentially, I found that there was just so much that I needed to learn, and it was going to be impossible for me to learn all of it. But my way of learning is through teaching.

And so I found that kind-of providing these monthly machine learning meet-ups where I would kind-of focus on a topic and kind-of have nice discussion around it, and kind-of solidify my understanding of it. Because there’s so much, I would rather focus on things that are very relevant and that are going to help me do my job, but understand it thoroughly, than know … like know a lot of things but at a very surface level. So that’s how I’ve kind-of gotten around it.

And I encourage people to teach others, because … and mentoring others. I think it’s a huge part of me and a huge part of what I believe is going to … It doesn’t only benefit the receiving end. It benefits me as well, right? So it’s a mutual benefit whenever you go through and you take the time to teach people.

Mike Delgado: And Sarah, and you’re doing a fantastic job with the work you’re doing at USC and these meet-ups that you’re doing. And I think that the education that you’re sharing on LinkedIn is awesome. And again, for those listening to the podcast or watching the show, make sure to follow Sarah on LinkedIn, because you will learn a lot just from what she’s sharing. And she’s sharing relevant things, things that are topical. So make sure you check that out.

We got a question here on Facebook Live, Sarah, around … Sandeep is asking, you know, how do I get started in data science? And it’s probably a question, I know, that you probably get a lot. And he says that … or Sandeep days, “I have poor academic records. How do I start?” And what would you say to the person who maybe struggled in school, but they do have a love for data science. They’re looking to break in. Where would you direct them to start their learning process?

Sarah Nooravi: So I think that the best way to answer this question is to kind-of approach it from the most general, the general case. Right? So say you want to transition into any career. Right? So what would you do? And what I was taught was that you go and you shadow someone who’s doing that job. Right? And you talk to ask many people who are in the field, and see whether or not you actually like what they do. Right? So I actually thought I wanted to go into … I forget. Accounting or something. And then I did an accounting internship, and I hated it. I was like, what am I doing here? This is actually not what I expected. Right? And so, talking to as many people as you possible can and really understanding what it is you enjoy about being in data science, and then using that to catapult yourself into kind-of working with real world data sets.

I mean, I would just formulate a problem, one that’s relevant to you, that you find exciting. Find a data set that aligns with it, or find a way to collect data on it. And then come up with some cool … I mean, before you can come up with some cool, maybe modeling approach, or some sort of tool that kind-of attacks that problem, I mean, do your research on exactly what you need. Right?

Don’t go too broad, ’cause you’ll drown in so much information. Just focus on exactly what you need to solve that one problem. And then refine it. And that’s, I think, the best way that you can start. And then find other projects that you find that are interesting and solve those problems one by one.

And then put those things in a repository, right? So keep all your code on GitHub. Make a GitHub. Put everything there, and make sure that, you know, you can dig back into your … to the code that you’ve written maybe a month, a year ago. Show it to future employers. Hey, look at what I did. But then also have that code so that you can reference it later if you need it. So, I think I would start there.

Mike Delgado: Sandeep, just take those words to heart. Keep pushing, keep learning, and just follow what Sarah said about courses you can, but yeah. Show some grit and stay curious and keep pursuing that, because that … It’s an exciting field. So before we go, Sarah, is there any last misconception you’d like to talk about?

Sarah Nooravi: I think one more that I can mention is that … I think probably the biggest one, bigger than the other two that I mentioned, is that as a data scientist, you are going to be modeling all day. And so, really the key to understanding why this isn’t true is that most companies really haven’t built up the data infrastructure to fully leverage you to that capacity yet.

So instead, you’ll find … you’ll be probably wearing a lot of different hats. So, that could be anything from doing ad hoc reporting, building Tableau dashboards, communicating your analysis to PMs, helping with the hiring process. So because you’re a part of this data team and you are the data expert, you are kind-of to be expected that you’ll be having to do a lot of these things.

And as a part of all of those hats, you’re going to be whether … I know it’s not the most exciting part, but you’re going to probably be expected to help with QA. And we all … No data scientist likes to spend their time QAing data. But you’ll probably be responsible of being a … ’cause you are a part of this data pipeline, right? And we don’t think or ourselves as being a part of this data pipeline, but we have to, as we’re … ’cause we are the ones touching the data lot, that we’re going to be expected to kind-of, when we see things that are incorrect or not matching up, that we go back to our data engineers and we let them know. So that way the data can get better over time.

And then even when we get to the modeling aspect of it, right? Time goes into framing the problem, understanding the data and the domain. I’ve been working at MobilityWare for three months now. And there’s so much data there. There’s the monetization, the marketing, and the product side. There’s so much data, and they’re all coming from different sources. So really, even three months in, I’m still learning about the data that we have, because there’s so much of it. And understand the domain that you’re in. Right? So I’m going into gaming for the first time, so understanding the mentality that goes into how people play games, what we need to focus on when we’re creating features. So that goes into kind-of this process of, okay.

Well, I want to be modeling, but I need to go through this process of framing the question, understanding the data, cleaning it and processing it, and then getting to the modeling. So like, it’s good to have …

But that’s the thing, right? Like, we have to know and understand and be exposed to all these different approaches that we could possibly take. But the reality of the fact that it’s going to be a fraction of a fraction of our time in our daily job. So …

Mike Delgado: Yeah. I read a stat, Sarah, and tell me if this is right or sounds right to you, that like 80 percent of the job of a data scientist is that data wrangling, the cleansing process. Is that … Does that sound about right to you?

Sarah Nooravi: Yeah. I would say. So, I think we … And that’s actually another thing that is kind-of unfortunate, is that we’re not really taught how to go about this data wrangling, pre-processing steps. Right? So anytime that we’re taught, either through … whether it’s a boot camp or even in our undergrad or wherever it is, or even in Kaggle when we’re approaching these projects that they have, we are pretty much given a relatively clean data set.

We download a .csv and then we just do the modeling part. So that’s great. We’ve been focusing on that part. But in the real world, you’re dealing with very, very messy data. And so, that’s another misconception, right? That we think the data is perfect and we’re just going to kind-of be able to just go right into modeling. And the reality is not that. And because we’re not taught that, I think it’s something that is an extra challenge for people in the data analytics, data science realm, is that we kind-of have to …

I did a whole meet-up on just data pre-processing, and I barely covered all the different techniques, like, to go about, like … It took me … like, you can look at kind-of the presentation that I put together. I was like, okay. I’m barely scratching the surface of all the different kind-of nuances to the data that you might be dealing with. So, because like we say, garbage in, garbage out. Right? So if you don’t have good data, no matter what model you use, it’s going to be meaningless because the data to start with is not in good shape.

Mike Delgado: Well Sarah, one last question before we end today’s show. And this is … You talked about earlier the importance of diversity both in domain knowledge, just because it makes a team stronger. It can help figure out and be more creative with solving data issues and data problems. What advice would you give to women within tech, specifically? Because gender diversity is just as important.

Sarah Nooravi: That’s right.

Mike Delgado: And I would just kind-of love to hear your thoughts on that.

Sarah Nooravi: So the best thing that I can say here is, don’t give up. Don’t let anyone convince, push, or bully you out. And this works … this advice goes to anyone who feels like they’re not being accepted into the data science community. I mean, it could be gender, but it could also be ageism. Right? So people discriminate on age all the time. They’re like, they feel threatened, right? Like, so what I would say is, don’t be intimidated. If this is something that you are passionate about, then focus on yourself. Focus on improving yourself. And you know, you might develop a little bit of thick skin along the way.

But even just try and find … So, one way that I’ve coped with this … I mean, I’ve dealt with a lot of challenges myself in this … with this issue. But I think my way of coping with it is just finding people outside of the data science team that I could be friends with in the company that I’m like, okay. So, I still feel a part of this community, a part of this, you know, bigger mission that we’re trying to achieve. Whether or not I find that directly, you know, with specific people that are working on my team or not, I still need to find a way to be grounded.

Right? And so finding those people that you can … that can help you feel grounded.And then being true to yourself. Like, I think that’s something that I’ve kep with me this whole time, is that even if I was the only girl in the class, I was like, hey. I’m not doing this for any of you guys. And actually, I’m going to prove you all wrong. Like, you may think something of me when I entered this class, but hopefully by the end of it, we’ll all respect each other. Right? Because I’m going to work just as hard, if not harder, and I’m going to prove all of, you know, whatever preconceived notions you had of me from the beginning, your … I’m going to prove you wrong. And so, yeah. I guess the best thing is, just don’t give up.

Mike Delgado: Great advice. Sarah, for those that want to connect with you, learn more about you, where should they go?

Sarah Nooravi: I Think the best place for me right now is LinkedIn, so if you connect with me on there. I’m not always super responsive, just because there’s so many, you know, so many messages that come in. But know that I read all of them, and sometimes during my weekends I blast through and respond to messages. But yeah. Feel free to connect with me on LinkedIn. I don’t know if there’s really a better place than that.

Mike Delgado: For those listening to the podcast, her name is Sara Nooravi, spelled S-A-R-A-H, last name N-O-O-R-A-V-I. So look her up on LinkedIn. And if you’d like to get a direct link to her profile, we have this entire transcript and video and podcast on our Experian blog. And the short URL is just And that’s, and we’ll have links to Sarah’s LinkedIn profile so you can go right over there, connect with her, or follow her. And like I said, she’s offering great advice. I’m sure you get inundated with messages, just because you’re so popular and active on LinkedIn.

So just, like you said, be patient. She reads your comments. But just engage with her in her posts, and that’ll be probably the best way for her to see your interest and … in the subject of data science. So Sarah, thank you again for being a wonderful guest in our Data Talk this week. And looking forward to hopefully chatting with you again in the future.

Sarah Nooravi: Awesome. Thank you so much for having me, Mike.

Mike Delgado: Thanks. Take care, everybody.

About Sarah Nooravi

Sarah Nooravi is a lifelong learner and data geek. She has a history of delivering innovative marketing tools to help drive better business decisions in the entertainment and gaming industries at Operam and MobilityWare. She is also passionate about teaching and giving back to the community. In that spirit, she is teaching a DataViz Bootcamp through USC, she leads and coordinates monthly Machine Learning meetups at Ticketmaster, and she mentors aspiring data scientists and engineers. All of these activities support a core motivation for Sarah: helping set up others for success in industry. Follow her on LinkedIn.

Check out our upcoming data science live video chats.

Never miss a blog post!

Subscribe to keep up with all things Experian.