Bias in Machine Learning models is a real problem, but also is the lack of representation of women and minorities in the more general Data Science field. Join me this week as I speak with Jessica Meyer about going from the start of her career working at Best Buy in their Marketing department to today being a Principal Data Scientist at Optum.
Jessica shares with us both her story and her passion for helping women and minorities be represented in the field of Machine Learning and Data Science. She hosts a popular podcast called Women in Technology Twin Cities and speaks at conferences such as the Women in Analytics & Data Science Virtual Conference this past fall. She is also an Adjunct Faculty Associate at Columbia University working in the School of Professional Studies Analytics Program.
If you are interested in learning about how AI is being applied across multiple industries, be sure to join us at a future AppliedAI Monthly meetup and help support us so we can make future Emerging Technologies North non-profit events!
Resources and Topics Mentioned in this Episode
Jessica Meyer 0:00
I'm often the only woman in the room a lot of times. So there's definitely a need to get more women and minorities in the space because technology plays such a huge role in our lives. And if we're building technology that doesn't work for everybody, that's a huge myth. And you know, I think we'll be in the same boat we're in right now with you know, having an economy that doesn't work for everybody.
AI Announcer 0:22
Welcome to the conversations on applied AI podcast where Justin Grumman's and the team at emerging technologies North talk with experts in the fields of artificial intelligence and deep learning. In each episode, we cut through the hype and dive into how these technologies are being applied to real world problems today. We hope that you find this episode educational and applicable to your industry and connect with us to learn more about our organization at applied ai.mn. Enjoy.
Justin Grammens 0:52
Welcome everyone to the conversations on applied AI podcast. Today we have Jessica Meyer. Jessica is a passionate data scientist with over seven years of experience in data analytics, data architecture, modeling algorithms, data modeling, data warehousing, and distributed computing frameworks. She's self proclaimed to have started her stint in data science as a search business analyst at Best Buy Corp, and has steadily progressed to her current position as a principal data scientist at optim, where she leads in the design and implementation of machine learning use cases and delivering end to end solutions. She is also an adjunct faculty associate at Columbia University working in the School of Professional Studies analytics program. And he's also the co host of the women in technologies Twin Cities podcast. So you certainly have a lot going on Jessica. And I'm very thrilled you've been able to make the time to join us this week. Thanks again for being here. Yeah, thanks for having me on. And yeah, I certainly like to keep myself busy. Since I'm no longer in a master's program. I felt like I had plenty of time now to spend another thing. So trying to use it wisely. No PhD in your future. I'm thinking about it. But those student loans, they can add up very quickly.
For sure. Well, you know, I talked a little bit about where you're at today, and some of the really interesting, cool stuff that you're working on. You want to maybe sort of like connect the dots, I guess you know, how you went from the work you're doing at BestBuy or even earlier if you want to just kind of the trajectory of your career?
Jessica Meyer 2:17
Yeah. So you know, one thing that and you mentioned that I do a podcast, one thing I've learned is the importance of telling stories, and especially a lot of people think that they don't have a story that others would find interesting. So I like to kind of go back a little further than most people would in my timeline, if you don't mind. I grew up in Minnesota, I've been here my whole life, I was raised by a single mom, we were very poor growing up up until probably I was like 10 years old. And we were on welfare. We lived in section eight, we had food stamps, we got government food with the cheese, which was like the best but the peanut butter was solid. It melted first for good use it. So I didn't have a whole lot of like technology interactions growing up. Yes, we had, you know, video games, but they were everything was usually like a hand me down and a decade old. But it wasn't till I was probably about 10 years old that my mom finally had fallen some stable income. And we're able to kind of come up out of the poverty level. And we got our first computer. And so I would spend time playing doom and Wolfenstein 3d, which I think growing up girls didn't put play on computers and play those games, but I loved it. I spent a lot of time on that. And unfortunately, when I was a teenager, I became a mom. And so that really completely changed the trajectory of my life, I had thought that I was going to you know, leave the state go to college, you know, have this completely different life. And all of that changed in a very short amount of time. And it was a lot to try to process. But I had a lot of good support from our family, and especially my grandmother, she was very instrumental. You know, I was very upset about becoming a parent at such a young age. And she said, You know, I love you no matter what. And I said, you know, that's right, it didn't matter what I did, I could totally fix this. Things completely changed your own. From there, I finished high school, I got an Associate of Arts degree, and still not really in the technology space. I always thought you know, I'm way more creative than that, like, nothing creative about technology, right? That's what
Justin Grammens 4:15
people think. I think that's a big misconception.
Jessica Meyer 4:17
Yeah, but there's kind of always in looking back, there was always this underlying thing with numbers and liking to compute. Like, I feel like I'm on my third career now as a data scientist. But you know, I worked in retail for almost a decade. And what I liked about it was that you had all these numbers and sales numbers that you had to crunch every day, and you had to try and figure out ways that you could make those sales numbers get better. And you were tracked and rewarded by the better you did. So like I would spend all my free time like crunching numbers and trying to figure out if I did this or if I sold this instead, you know how that would change things, and not really realizing you know what an impact that would have later on. But after the recession, I was laid off and I had the opportunity to go back to school and finished my four year degree And I thought I wanted to be in marketing, because that was super creative. And I ended up at BestBuy not in a marketing role at all, but just doing some, it's actually for holiday and it was looking at data and, you know, making decisions of whether you change the results of the search algorithm. And so that what I found that extremely interesting, and I loved Best Buy, because I had been, you know, since a teenager, every Tuesday going there to see the latest movies and music and video games that were coming out. Right, that was the thing my day, and I really wanted to stay there. And so I kept finding ways to bring more value to the role and being on search in them, you know, kind of transforming their technology, I got the opportunity to really work on a software development team, and understand how you develop software. And that's where I got not only interested in, in data science, but also software development in general and building things. I didn't think that was something I would ever be interested in. But just getting to sit down and work with developers on a daily basis and see how they worked really inspired me to be able to do that same thing. At this point,
Justin Grammens 6:02
what was your interaction with the software developers? Because you weren't writing code yet? Right? Or were you getting the opportunity to do some of that?
Jessica Meyer 6:09
So I wasn't writing code, what I was doing is I would look at the kind of the results that the code was returning in search, and then looking at the data, and trying to understand how our relevancy algorithm was performing. And if we made you know, these changes, would that perform better than doing all the testing and analysis for that? And that wouldn't require me though, sometimes to be running, you know, things in command line, but they weren't things I came up with. They were, you know, sub steps that I was given by a developer to do so.
Justin Grammens 6:37
Okay. You mentioned the recession, there's been a couple of them. So is this is this early 2000s? Or is this the 2008? Nine recession?
Jessica Meyer 6:44
That would be the eight nine recession? Yeah. Okay.
Justin Grammens 6:47
And so you're at BestBuy, you made it through that recession? I guess, at Best Buy, were you there before, during after
Jessica Meyer 6:53
I was there a little bit after? Yeah, I was working in furniture at the time. And so when the housing market burst, people were not buying furniture. And so I got laid off. And of course, you know, at the time, they were extending unemployment benefits and things like that. So I was given the opportunity to just go full steam ahead and finish my bachelor's degree. I was this was around like, 2011. And you know, people were really struggling to find jobs coming out of college, then, even though things were getting better, it was still hard to find jobs. And I was working with a lot of, you know, different contract agencies in the Twin Cities. And so I was fortunate to not only land at Best Buy, but land in the technology side of the house, because at that time, they were really struggling financially with Amazon's competition ramping up and some some other things happening. So even though, you know, the stores were struggling online was growing. And so they were willing to make investments in the online business.
Justin Grammens 7:47
Yeah, for sure. I mean, I was at Best Buy around that time to where they contracted our company, one of my first companies recursive awesome to build a lot of their initial mobile apps. So it was an interesting time, because you're right, Amazon was definitely eating their lunch. That's why I was going through I feel like a big transformation of that time, but they, you know, credit to them, they were doubling down with regards to mobile development and even calm, they were just trying to make the best experience that was possible. So I was fortunate enough to help them build out a lot of some of that some of these feasts, these first online experiences in iOS and Android.
Jessica Meyer 8:20
Yeah, it was definitely, I think, a rewarding time. Because just to see a company go from being in kind of a dire situation to fully turning the ship around and to see where they are. And knowing that I was a part of that. It's really something to be proud of.
Justin Grammens 8:33
Yeah, for sure. So you're there and you're noticing you're getting exposed to technology, you're probably getting a little more excited about this. This is cool. That was it around then that you were like, Well, let me go for a master's degree or a data science maybe wasn't even a thing, you know, then or I guess, what were you? What were you thinking about at that time? And where were you headed?
Jessica Meyer 8:49
Yep. So I was always, you know, my job was literally how do you make the search results better? How do you order things in a way that customers are more likely to purchase something. And so I was always trying to find ways to make it better. And I knew that that would take more data, and different types of data and different ways of processing it. And so what I was seeing is that, you know, you look at what Google and things were doing with their search, those were kind of like the inspiration of, you know, what I was trying to get at, and, you know, they were very heavily in the data science space. And so, you know, I had one of our directors, I'm sure who we both know, a rune Batu, who brought in the dean for the data science program at, you know, at St. Thomas and kind of gave a rundown of what that look like, then that's where I got, I think I can do this. I just need to learn more, and I'm more of the hands on but I also like, kind of the classroom approach. I need that to be held accountable in order for me to like further pursue things.
Justin Grammens 9:46
Yeah. So it's like, Okay, I'm gonna I'm going to go I'm going to jump in here, you know, jump in both feet and go back to school. I went through the St. Thomas program as well while I was working, so I kind of know that experience with regards to having to work During the day and take night classes, and it can be those can be long days, and many, many, many, many years. I mean, my I took I think I took for at least four years for me to finish mine.
Jessica Meyer 10:09
Yeah, that's how long it took me to I think I started in 2014 and finished in 2018.
Justin Grammens 10:13
And so you came out of the program, but you continued then to stay at BestBuy. Right? And you did you? were you doing data science stuff there at Best Buy? Yeah, so
Jessica Meyer 10:23
what was kind of nice, and I really appreciate the leadership support there that I had, you know, they were always willing to, like, work with me and trying to find opportunities for me to use what I was learning, especially they were kind of new to data science as well, you know, I think there was maybe one team that was using it, like in the.com space. And so I had opportunities to, you know, talk about data science and help find use cases and help make decisions about what tools that they should be using for you know, to create this, you know, Center of Excellence, and get those conversations started. Although my skills were kind of developing quicker than the opportunities were. So I ended up wanting to I did some time in analytics and looking kind of at, you know, marketing type problems. But I knew I wanted to get back into the like deploying things and to production. That's kind of what my life had been in what I enjoyed doing was putting things into production that I had been a part of whether I built them myself or not. So that's how I ended up at optim is there was an opportunity to build out some things and you know, deploy them in production environments.
Justin Grammens 11:28
So what do you do then on a day to day basis, like, what's a day in the life of a principal data scientist at often these days,
Jessica Meyer 11:34
it's very busy. We have a lot of low hanging fruit. And I, you know, there's different types of data scientists I fall into, there's ones that are, you know, more on the research side of things where you're coming up with insights. And I'm more on the technical side. So the insights, while they are kind of important while you're building the models, and building the use cases, I'm more into implementing the machine learning and kind of marrying them with technology so that we can make products that are smarter and work better. I'm kind of really into right now, how to look at infrastructure and make sure that it's stable, and that when problems occur, you can find them quickly and resolve them quickly. So that's kind of the space that I've been in.
Justin Grammens 12:14
Is it kind of under that ml ops? umbrella?
Jessica Meyer 12:17
Yeah, to an extent. Yep. Good. Okay.
Justin Grammens 12:19
Jessica Meyer 12:20
high volumes of data.
Justin Grammens 12:22
I'm sure there's a lot. And so are you are you looking at? Well, I saw a couple things, I guess, come to mind. I mean, is is it has COVID impacted with regards to just you said things are busy, we're just things we're kind of saying things are always busy, or are they a little bit more busy now, because of everything that's going on with the pandemic or things like that? Are you kind of largely unaffected by that?
Jessica Meyer 12:41
You know, we're always busy, especially we, you know, being in healthcare and the applications, we're supporting our healthcare related, we've definitely gotten busier and had to act quicker than we had in the past. Yeah, I mean, it has made us busier. But, you know, we've kind of largely been running full speed ahead to support some of these, because healthcare is so important to so many people. So,
Justin Grammens 13:04
yeah, for sure, for sure. So do you find yourself getting deep into a lot of the code these days? Are you writing Python and doing data cleaning and sort of end to end? Or is it really more? Well, obviously, you're leading a team, so you have a bunch of other people that can help you, but where's your focus most of the time?
Jessica Meyer 13:19
So it's important to me that I have at least one project where I'm writing code. Yeah, I definitely do probably more consulting than they do writing code these days. You know, we have data, you know, some Junior data scientists that are also building and, you know, learning how to how to implement these projects at the same time. So I do spend a lot of time with them, and helping them you know, and making sure that they're empowered to make their own decisions. And, you know, teaching them kind of the process of how to engage with people. So that does consume a lot of my time. But I'm also like, aware of everything that's going on in the team at the same time, which I find interesting.
Justin Grammens 13:53
Good, yeah, no, I mean, I've gone through various phases of my career, where I'm out of the code, and then I'm back in the code, and then I'm running a team. And then I'm, you know, back in the code again, and I'm just an individual contributor working on things. And personally, I like to mix it up a lot, especially in new areas of technology and sort of being able to, in some ways, I think, if you're going to lead a team, you got to kind of understand what they're going through, in some ways, you have to kind of drink the Kool Aid or whatever you want to say before, then you can start leading a team around all the aspects of it.
Jessica Meyer 14:22
I'm fortunate that we do have a director of data science, who's more of like the management side, so I don't have to do a lot of the like, people management side of it, but more of the like, you need help with the technical stuff. And, you know, just sometimes when you're trying to figure out a problem, even if the person doesn't have the answer for you, just talking through helps you find the answer. And I get to play that kind of supporting role, which is nice.
Justin Grammens 14:44
Well, yeah, the The other thing that you have done is becoming an adjunct faculty associate. You didn't do that while you were in school. But I think once you got done with that, then you decided to, I guess, give back in a way, you know, help people by becoming a faculty associate. How's that process? Ben?
Jessica Meyer 14:58
Yeah. You know, it's funny. So when I was a kid, I used to play school all the time. So I was the teacher as a kid. And I'd have my room be complete mess, but each student would have their own desk, and they're all paperwork. So it's always been kind of a passion of mine and something I wanted to do. It's a lot of work, but not really, because it's mostly, you know, sitting in on class and making sure that as a teacher teaching assistant, I'm not, you know, giving the lectures, but I'm helping the professor with the questions and, and making sure that things are going smoothly for the students. And I get to, you know, meet people that are just starting out on coding, and they're super nervous. And I can totally relate to that and talk them down from the ledge and get them back on track and get things working.
Justin Grammens 15:42
That's awesome. Very cool. Very good. Well, I, one of the things I like to ask people, when they come on the program here is, you know, how would you define AI? You know, if we do have a elevator pitch or something, that people if they either ask you about that, or even define your job, like, you know, like, what are you doing on a day to day basis? Because I think people hear the term, and it just sounds scary, or just something that's futuristic, or, you know, that just doesn't make a whole lot of sense. So I sometimes just like that, you know, ask people if they have a just a general definition, or what they think about when I think about AI and machine learning?
Jessica Meyer 16:13
Yeah, that's a great question. I think one I often struggle with too, because it changes so much. And even just the definition between data science, machine learning and AI, I feel like right now, it's all kind of being blended into one thing, but I definitely see how it's separate. Like, I like to think of data science is more just doing science on data, like, you know, using statistical methods to understand what's happening in the data. And then I look at machine learning as more of using the insights you learn from data science to train the machines to make a decision. And then when you when I see AI, I think of more of the machines are making those decisions, and they're learning on their own how to do that. And that's kind of how I delineate between the different explanations or definitions
Justin Grammens 16:54
of them. Excellent. Yeah. Makes a lot of sense. Makes a lot of sense for sure. And as people I mentioned, about people getting scared, or there's there's fear, do you have any sense of AI machine learning this stuff kind of taking over jobs? You know, I there's some people that I talked to who are concerned about the future of work for humans, like, how are humans going to play out once this thing gets to the end? And I guess there's no end per se, but as it keeps moving more and more into this computers doing work for themselves? Because they can figure things out? Do you have any thoughts on that?
Jessica Meyer 17:23
Yeah, so I think we've kind of been through this cycle before, with machines taking over our jobs, and maybe look at how we've gone from being farmers to the industrial, you know, to even I mean, looking at the examples, the car manufacturers, that was a big change, you needed less people to do that. But I feel like it frees us up to look at better, more difficult problems to solve. So I don't see it taking over everybody's jobs and us not having something to do. I mean, there's a whole world who just landed on Mars again. Yeah, so we have a whole world of problems that we need to solve that AI machine, if we can't figure them out, machines are gonna struggle to figure them out. So I think it just frees us up for that, you know, the higher thinking and the harder problems.
Justin Grammens 18:08
Yeah, that makes a lot of sense. As it kind of talking to you about your background and stuff like that kind of what you've gone through and your entire career going up to where you are today. What do you think is your greatest strengths and weaknesses as you're then now, I guess, thrust into this third career that you're talking about? What are some some things that he as you look back at yourself, that you might find yourself as a strength or a weakness,
Jessica Meyer 18:30
I'm always trying to learn new things and get better at things. I definitely a continuous learner. And that's probably been my number one strength, you know, even as a kid, you know, reading books, and being inspired by others and seeing what was possible. Because I think with the background that I have, like, you know, my parents didn't go to college, I'm the first in the family to get a college degree. How do you know what to aspire to? If you can't look around and learn from what other people are doing? You know, outside of your personal bubble? As I would say, that would be my biggest strength.
Justin Grammens 19:01
Sounds good. Are there any applications of AI in the world? I guess, that you've seen recently that you're like, Oh, this is this is really interesting, you know, and it could be within optimum within healthcare. Just even, you know, I'm not sure if it's something you article you read recently, or somebody else had a conversation with. I'm just always curious to see people that are practitioners in the field. If there's any things that have like, made them say, wow, I don't know how that's done. That's kind of blows me away a little bit. I wasn't sure if you had anything.
Jessica Meyer 19:30
Well, I'm still like stuck in autonomous driving. I drive a Tesla and like, I'm very big proponent of that just the potential to reduce accidents and saving lives. Like it's huge to me, so I'm always following along with, you know, where that is, as far as like when will it'll hit the road and stuff. And I'm also obviously I'm a huge Elan musk fan. So like the neuro link stuff I'm definitely following along. I grew up watching Star Trek. So the idea of like having your brain downloaded in a computer and you know, live on forever very appealing to me. So definitely paying attention to what how that would work. And some of the things they're talking about, you know, using computers to, you know, heal people and things like that. So those are kind of been my focus.
Justin Grammens 20:12
Are you good? Yeah. Do you have any you mentioned Star Trek, you have a favorite book, you know, either AI or non AI related, any things that you've found as you've been working through your career that you found interesting?
Jessica Meyer 20:24
Yeah, it's funny, because I'm like, one of my side passions, I guess, is diversity, inclusion and equity. So a lot of the things that I'm doing outside work are kind of around those. So I'm always reading books, like, I really loved Melinda Gates, the moment of lift. And, you know, Michelle Obama's becoming like, I'm always interested in other people's stories and how they got to where they are. So that's kind of where I focus my time, when it comes to, you know, work stuff like AI, I have a Google News alert, so I can kind of consume that quickly. I think there's a lot of information out there, especially in the space that people are really interested in, it can be really hard to consume it and not be overwhelmed. So I'm very intentional about you know, how I consume things that are related to work and kind of trying to balance it with my passions.
Justin Grammens 21:10
For sure. You know, I said at the beginning that you have a podcast called the women technologies, Twin Cities podcast. How was that? Ben, you want to share a little bit about maybe what happens on that podcast and what your mission and some of the presenters you had on that?
Jessica Meyer 21:25
Yeah, so I've spent probably five or more years of my career, you know, being involved in employee resource groups that are kind of focused in specifically in women, and diversity and inclusion. And, you know, when you look around at Tech at the technology landscape, I'm often the only woman in the room a lot of times, so there's definitely a need to get more women and minorities in this space. Because technology plays such a huge role in our lives. And if we're building technology, that doesn't work for everybody, you know, that's a huge myth. And, you know, I think we'll be in the same boat we're in right now with, you know, having an economy that doesn't work for everybody. So I'm very passionate about bringing more people into the fold. And that's kind of what we tried to do with women in technology with the idea that people stories mattered, how they get into the roles matters, because it will inspire the next generation coming up. And so that's what we focus on is just meeting women that work within the Twin Cities community that work in technology and having them tell their story and how they got into it, because it's not a linear path by any means. And we we all take very curvy, sometimes uphill, fast to get to where we are. And so being more transparent about that will help the people that come along behind us.
Justin Grammens 22:39
That's great. Yeah, I i've been actually trying to bring more women and just diversity on this podcast, quite frankly, because you're right, as I reach out and start talking to people that are working in this area of machine learning, and AI typically is, you know, male, because it's just it's technology. Same thing with software with like the software teams that I lead and stuff like that. So I appreciate you being on the program, just to give that perspective. And I really, I respect what you're trying to do just with the podcast, and just getting more and more women to tell their stories and kind of like, let other young women understand that this can be done, right, the more women they can see, as role models, the more I think opportunity, then they're gonna want to jump in themselves.
Jessica Meyer 23:19
That we do, we did an episode that was kind of interesting about the history of women in technology. And in software development. A lot of people don't realize that women were actually the first programmers, because they thought that the computers were huge. And it was such a remedial task that they left it to the women. It wasn't until they figured out that they could be making money off of software programming and computers, the kind of the men took over. And they had, like, two psychologists that did this study that was really biased to figure out who the best software programmer would be. And that's kind of been the basis for hiring for the last 2030. Some years, they actually said they wanted a male who hates people was kind of like, what they were looking for someone who doesn't like people and who was really analytical and into mathematics. So it's kind of interesting how it's shaped to the industry.
Justin Grammens 24:13
But yeah, I mean, that's almost like a self fulfilling prophecy, right? I mean, it's just like, if those are the type of people then that is the view of a programmer just introverted, that sits there and just does things by themselves, which as we both know, that's not the best type of person you want to have on your team.
Jessica Meyer 24:29
Yeah, definitely a collaborative protein, you just build better products when you involve more people and get more input. So yeah, it was definitely kind of an eye opening episode for us to figure out how we got here, because you hear a lot about the pipeline and it being linky. And, you know, there's a huge shift that happened in the 80s, where you had, you know, a lot of women in software engineering programs in college, and then there's just this huge drop off and trying to understand why that happened. And that would be why,
Justin Grammens 24:56
why didn't I know about that? Well, as you were talking about it, Some of this inclusion stuff, I started thinking about just AI. And there's a whole ethical AI movement that's going on right now. That's the kind of a hot buzzword right now. You know, it's kind of off the cuff, but I did it. Did you guys or have you, you know, on your podcast? have you covered some of that stuff? Talking about that kind of the cross? Or have you read much about that? Are you interested in that?
Jessica Meyer 25:20
I'm definitely interested in it. Because I mean, there's so many we've talked about the examples of when it goes wrong. You know, for instance, I can't remember which, if it was Snapchat, or there was one filter that would change your face. And it did not pick up people who were dark skin. And just because they didn't have enough samples, you know, their algorithm, so that, we've definitely talked about examples where it goes wrong. Same thing with like, we don't think about this too often. But when you go to get Soap Dispenser, or an automatic paper towel roll, if your skin is dark, it has a hard time picking that up. And, you know, saw even watched videos of people with dark skin, pulling manually it out and then putting the paper towel underneath with their hand so that they can get more out or get the scope out. So we definitely talk about examples where it goes wrong, and why it's important, but definitely an area for us to explore for sure of how to reduce that bias and be more ethical and what we're developing, because it will be very important here in the future.
Justin Grammens 26:18
Yeah, for sure. I mean, I'm not sure if it's 2050 2060, or something like that. But you know, the Caucasian population is going to be the minority, actually, in the world, will actually probably already is with regards to China. But I mean, just even in the United States, you know, more and more people are gonna have different perspectives, we're gonna need to account for it and coming bring it back to machine learning and AI Yeah, for those algorithms can't handle that you're gonna have a broken system, no one's gonna pay money for it,
Jessica Meyer 26:42
especially in health care. In particular, there's a competition going on right now, with a datathon, with the Women in Data Science group out of Stanford, the problem that they're trying to solve is related to COVID. And that, if you're diagnosed with diabetes, that makes a huge difference in the care that you receive. And a lot of times these patients are coming in either unconscious, or they don't, and they don't have a history, a medical history for them. So they have no idea, you know, if they have diabetes or not, or if they're at risk for it. And so they want to be able to predict that. And when you look at kind of the data set that you're given, you see that a lot of the data they have is based on Caucasians that have diabetes, and that's not the percentage, you know, people who have it isn't necessarily Caucasian. If you were to do some of the data cleansing techniques, you know, like taking the average, what would you fill in all those missing values with it would be Caucasian, and that would bias your algorithm, then? Yeah, I
Justin Grammens 27:35
had read something, there was an article, I think it was Amazon was using machine learning. And I was just talking to somebody about this the other day where they were using a filter based on resumes, you know, they must get 1000s of resumes a day. So they were using a filter, but the filter was learned based on the prior 10 years of resumes coming in. And the people that were applying, were typically males, actually, it's kind of going back to earlier on in the conversation. And so when the resumes were actually once at once with this filter, and the people were actually now humans were reviewing them, it was completely filtered out all the women, it completely filtered out all the names and people that you know, not the John's and the jeans and the oil, you know, Joe's or whatever, you know, it was completely biasing them feeding back on itself to now only hire the same white males that they have hired in the past.
Jessica Meyer 28:18
Yep, definitely an example, too, that we've kind of talked about, too, with Amazon's algorithm yet. One rule of thumb is, you're going to be in the news, it shouldn't be in the news about something like that. We should all be thinking, are we setting things up in the right way so that we're not in the news as one of those bad examples?
Justin Grammens 28:35
That's a good perspective to have, as you're working with abdomen? You know, obviously, you're dealing with a lot of different races and a lot of data just sort of across the board. And healthcare is a very important place to be on these days. So it's exciting. When it comes to doing it the right way, I guess. Are there any advice on classes or podcasts, blogs? Any any other stuff that you think that people should take? Or how would they start their career in this?
Jessica Meyer 29:03
Yeah, I get asked that a lot. It's a balance of things. I believe in going the, you know, the traditional route. That's what I did. That's what worked for me, but it doesn't work for everybody. And I don't think that it's necessary. There's so much so many resources online, especially if you're trying to learn to code. I think people always want to start with the coding part, and not the actual like methodology behind why you would clean data a certain way or choose this algorithm over another one. But I really think it starts with that. And so that's why I like the traditional college background because you do get that along with the coding depending on the program that you go to. And I also think having a mentors is really important. Someone that's working in the field and those those can be hard to find. I know I've myself struggled with finding mentorship within the data science community, but it's important because they teach you not just, you know, you want to make sure you're going to them for the right problems too as well. I think people a lot of times expect a mentor teach them everything, you know, the technical skills, the hard skills. And that's not really what's best, they teach you more of the soft skills and how to make better decisions and being a sounding board. So I think that's a definite mostess to find a mentor in that program. And then I feel like Coursera has good courses. Stanford is coming out with a lot of good courses that are, you know, online and either, you know, a minimal fee or free, I do a lot on data camp, at least in the beginning, when I was trying to learn how to code because they have both Python and R. And you know, that's a fee, like $25 a month, but totally worth it. And it actually really helped me a lot on my assignments. Once I get deeper in the program, those are kind of a few of the resources that I look to
Justin Grammens 30:42
nice. Have you gone through any, like the challenges, I guess, some of the data science challenges,
Jessica Meyer 30:47
you know, I'm very busy, so I don't get to do that. But that is also a good point that, you know, kaggle competitions and things like that are important. I also like attending conferences, just being you know, in the room with like minded people, even if you're not like absorbing the content that's being presented, I feel like it kind of inspires you to kind of recharge your thinking and get more creative. There's the mini analytics conference that happened throughout the year, those have been really amazing. And, you know, they're starting, they had their first women only group in October, I had to present it that and that was really interesting to see how many different women in the area and what they were presenting. So yeah, I definitely would look into those types of things as well.
Justin Grammens 31:29
Awesome, Jessica? Well, speaking of mentorship, and stuff like that, how can people reach out to you are you on LinkedIn?
Jessica Meyer 31:36
I am, I am on LinkedIn. And it's just Jeff Meyer, as the easiest way to find me, although there's like 100 or 1000 diamonds, or 1000s of results for finding me, but I'm the other way is through the podcast. So we have a website. It's with Twin Cities calm. There's a contact form there as well. And then on Instagram, it's wit podcast, I believe wit TC podcast. Cool.
Justin Grammens 32:01
Yeah. Are there any other things that I might have missed any topics or projects or anything like that, that you working on that I didn't cover that you'd like to share with our listeners?
Jessica Meyer 32:09
I you know, I think you covered everything. Now we've just started recording for our podcast, or coming out with season three, we added a new co host, her name is paid. And she's a lot younger generation just getting started. So I think she'll bring a fresh perspective to the technology community. And she's kind of new to the Twin Cities area as well. So we'll get to hear her perspectives, you know, from a fresh new face and get her insights into what she thinks that the community and
Justin Grammens 32:39
that's really cool. Really cool. Okay, definitely. Well, yeah, as I said, we will be sure to include links to the podcast and links off to ways to contact you as well. And, again, I appreciate you being on the program here and all the help that you're doing both becoming an, you know, a faculty associate at Columbia University, doing this podcast, obviously, you know, you've achieved a lot here along the way from when you started your story here at the beginning, when we opened this podcast, so I think it's gonna continue to improve, you know, I'm sure you will, you will. So, and again, super excited for you. And I'm really, really glad you've been on the show here, Jessica. Thank you again.
Jessica Meyer 33:18
Yeah, thanks for having me. It's been fun. Take care.
AI Announcer 33:22
You've listened to another episode of the conversations on applied AI podcast. We hope you are eager to learn more about applying artificial intelligence and deep learning within your organization. You can visit us at applied ai.mn to keep up to date on our events and connect with our amazing community. Please don't hesitate to reach out to Justin at applied ai.mn if you are interested in participating in a future episode. Thank you for listening
Transcribed by https://otter.ai