Forging a Career in Data Science

[00:00:03] Margaux: Welcome to UVA Data Points. I'm your host, Margaux Jacks. Interested in what a career in data science can look like? Today we're joined by two members of the School of Data Science Advisory board. Heidi Lanford, co-founder of NavAlytix AI and former Chief Data Officer at Fitch Group and Kane Guyer, principal at PwC and most recently the leader of the US and global chief Data Office. In conversation with Reggie Leonard from UVA School of Data Science. They share perspectives shaped by decades of experience leading data and AI initiatives across global organizations. [00:00:43] Reggie: Well, today we have Kane and Heidi with us from the board at the School of Data Science and this is amazing. I'm Reggie Leonard from the School of Data Science. Maybe we can give brief intros, and we can start with Heidi. [00:00:55] Heidi: All right, I'm Heidi Lanford, I am an alum of UVA, graduated from College of Arts and Sciences with a Math and statistics degree and I've been serving on the board for the past 11 years and right now I am the co-founder of an AI startup company called NavAlytix AI and we help companies with identification and implementation of their best AI use cases. [00:01:22] Reggie: That's perfect. And Kane, what brings you here today? [00:01:24] Kane: I'm Kane Guyer, I am a principal at PwC. I'm also a graduate, an undergrad degree in Environmental Science, class of 98. Very happy to be part of the board here. Now actually rounding out probably my first year and as I said, I've served in the advisory practice. I was an advisory professional at PwC for the first 20 years of my career and spent the last five as our global and US chief data officer. [00:01:53] Reggie: That's excellent. This is great. I'd love to kind of ask some follow up questions and I have a lot of questions here. But before we get into to a little bit more granularity in your backgrounds, what's a part of your story that helps explain how you became who you are professionally but that wouldn't necessarily show up in a resume. [00:02:11] Kane: So definitely something that's not gonna jump across My resume is embracing the culture at PwC quite frankly and my former employer who was bought by PwC diamond of taking 5 to 7 year inflection points and really I think reinventing yourself, taking risks, jumping into new roles, taking on new responsibilities, even moving from one sector into another, it was daunting. Sometimes the decisions that I made were definitely driven with a little bit of fear. But at the end of the day those inflection points were the areas I can pinpoint my career arc that I Grew the most when I was uncomfortable. I grew the most when I got complacent, started making mistakes, got a little bored and probably started looking for other opportunities, quite frankly. So that's not gonna jump out of my resume. [00:03:08] Reggie: Yeah. I'm gonna ask you a couple of quick follow up questions. So you mentioned being at Diamond. They were purchased by PwC. A lot of times when folks are navigating careers, a lot of the ways that people's career trajectories show up on LinkedIn isn't as obvious to them. And so a lot of times organizations acquire other organizations. And so I think it's interesting that you were at PwC through an acquisition. Is that a pretty normal common thing for large organizations? For folks that are figuring out kind of, should I go for the big name brand company or is it worth applying for smaller companies that are lesser known and things like that? Those are the types of questions that we get a lot of times from students. Is kind of this first job outside of school is deterministic of the rest of my career arc. [00:03:58] Kane: Yeah, I fundamentally don't believe that. I mean, straight out of the gate, I mean, I worked for the World Bank Group. I had to put food on the table. [00:04:08] Reggie: Right. [00:04:09] Kane: So I mean, there's kind of that tension. But once I got into something that I felt really good about, started building a little bit of confidence, I went to work for a company called Cluster. Cluster was bought by Diamond. And Cluster was very small, a very small North American footprint. Yeah, it connected me with the president of operations directly. I was sitting like right across from the office. So that was great. Diamond was, you know, 250, 300, I think 500 total. Headcount consulting company focused on management and technology. That intersection taught me a tremendous amount. I mean, everybody around me was an expert. An expert, you know, posited against a specific sector. I learned a tremendous amount. I could call up Mel, who was the CEO as a senior associate. That was great. I think traversing the gap is just kind of where you need to be mindful. And PwC did a phenomenal job. I mean, there are mergers and acquisitions that completely, you know, the cultures are different. Completely fail out of the gate. I mean, look, I'll be honest. I was spiffing up my cv and three months in I was like, no, this is going to be my home. And that was in 2010. [00:05:17] Reggie: That's awesome. [00:05:17] Kane: Wow. Sometimes there are things that you can control. You can always move right. If it's not a good fit, this just serendipity luck. You know, just happened to happen that I work with the same people that I've worked with for 25 years and I'm not going to get the gold watch of 25 years. But I'm happy, I'm so happy that that's. [00:05:36] Heidi: They give gold watches still. [00:05:37] Kane: No, that was an old. [00:05:39] Heidi: I should have stayed there. [00:05:41] Kane: Come on, I love it. [00:05:44] Reggie: You say you should have stayed there. Heidi, like you spent some time at. [00:05:47] Heidi: Well, I was at PwC, so it's interest. So I'm gonna skip your first question for now and riff off of Kane's. So I, coming out of university, I will say I felt like the big name would be really helpful to be a good launchpad for my career, wherever it would go. However, I caveated that with. I really needed to see that there was a mass, a group of people that were in the same discipline that I was, which was. We didn't call it data science back then and I'm older than this guy but, you know, I needed something other than to be a one off or a two off. And so I had, I had offers from Accenture to go into their practice and be a general management consultant and go to their boot camp for three or six months. But there was not a specific data science practice there. PwC. It was PwC at the time actually had the data science practice that none of the other big six had at the time. And so there was a mass of like 200 people there. And I really felt like I could learn and grow and develop. And so I went that path. And I did have some other offers with smaller companies in the D.C. metro area, but then I would have been, you know, one of two analytical statistical programmers. And that wasn't enough for me. [00:07:17] Reggie: And why is that? Were you looking for mentorship or. [00:07:21] Heidi: I wanted to learn a lot and I didn't want to just be the person in the basement with no Windows coding all day with a can of raid. And that's the thing that you have to analytical tech folks have to watch out for is being, you know, relegated to something like that. [00:07:43] Reggie: That makes sense. It's interesting that you use the word relegated in that sense because one of the things that I've seen is that folks as they progress throughout their data science career are navigating this individual contributor IC or management trajectory. And it seems like the most obvious trajectory is management. But if they still want to be an individual contributor, it feels a little bit harder to navigate that. But it sounds like you knew that you were interest in not necessarily only doing the IC type of work, you wanted to be exposed to some stuff. Is that fair or. [00:08:19] Heidi: I don't think I knew that coming when I was 22, coming out of college. But I will say I think you can have both. And I actually think it's a good thing to be really good and go deep in something and be that ic and then you can spread your wings and go a little bit broader. And that is the path that I chose. But one thing I would say that maybe doesn't stick out on my resume as much is I am so focused on the adoption of all the things that we build and create as data scientists, even more so than making those assets incredibly technically sound and rigorous. I feel like if you can get 70 or 80% there in terms of an awesome model or algorithm and you get like 100% adoption, that's way better than 100% awesome algorithm or model and 20% adoption. And so I've focused a lot on the human side of what's in it for me or the so what in it. And I feel like a lot of that actually started with my training here at UVA because some of you know my story. I got into the engineering school and I dropped out of the engineering school. I really didn't like it and I felt like I was suffocating and I wanted to get experience with all of those liberal arts things. And it wasn't because I wanted to learn how to be a good communicator and the role I'm in now, but I was really interested in those things. And I feel like that shaped how I think about data and analytics because it's still at the core. So I dropped out of engineering, became a math and stats major, took all the programming classes I could in the eschool and a lot of business school classes. Kind of made my own undergrad data science major when we didn't have that. But I still anchored on the math and stats and the applied aspect was the, I'll call it the wrapper around my skill sets. [00:10:33] Kane: Well, you moved into a pretty easy curriculum there. Math and stats. [00:10:37] Heidi: Yeah, no comment about environmental science. [00:10:44] Reggie: Tell us about that, Kane. I mean, I know that you're, you know, like a marine biologist. Could have been a different path for you. And in my world of career development, we have the theory of life design. There, all of there are multiple viable life paths that are true to you, that you could viably live. And you have to make a choice. And a lot of times that's really difficult for, for folks to navigate. Earlier in their career and definitely in their studies. And it seems like, you know, you had this marine biology life and then this chief data officer life. And so tell us about that, you know, two path diverging in the woods sort of story. [00:11:22] Kane: Well, I think so. My dad, you know, I listed, I think in some of my discussions with some of the students, I've always said that my parents are kind of one of my anchors in terms. I mean, they were just great mentors and, you know, they led by example. And so they were risk takers on one side. But my dad always said, you know, look, things are going to come at you. They're going to be things that you control, that you can't control. And, you know, sometimes you'll be on the left side of the line, sometimes you'll be on the right side of the line. But everything, if you take positive, you know, a positive outlook, you'll land on your feet. And, you know, I did. I wanted to be the American Jacques Cousteau. Failed. I wanted to be a neurosurgeon. I actually wanted to be a neurosurgeon. Coming in, I think I said that I was like, I'm going to be out here in three years, get the AP credit, I'm going to go right to med school and just be a brain surgeon. Failed. But I took bits and pieces of that through that curriculum, through UVA in the college I was taking molecular cell biology lab. And what I found is that when I get my sleeves rolled up and I'm working in either a lab environment or an outdoor context like I was in environmental science, collecting primary data, synthesizing it, and then trying to kind of make the next best decision on where we actually take it was thrill. And so I got this foundation of not necessarily the competency that Heidi, I think at least started to develop in undergrad, but I at least got a passion for it. I picked the best pieces out of that. And I said, in order to, I think, make this a career, I need to marry that with technology. And so in 98 to 2000, I Day traded to feed myself off of a little investment that I had. And I began kind of outside of normal university paths, learning about technology. And then in 2000, I got a job that was both data and tech. And it's kind of history from there. And so, you know, if you have a positive outlook, you can pick the best things. There are certainly a lot of life experiences that I would say, I wish I could do that over again. You can't. [00:13:34] Reggie: So, yeah, I feel like I would be remiss to ask how marine biology shows up in your life, though it's not a job title. I'm sure that hasn't left you. [00:13:44] Kane: No, it hasn't. I've got grand plans to build a, a dive tank in my backyard. [00:13:50] Reggie: Wow. [00:13:50] Kane: Just, you know, I haven't passed that through management, but I'll work on it. [00:13:54] Reggie: I hope this isn't the first time she's hearing it on the podcast. [00:13:57] Kane: It is, it is. She's got a 10 year girls trip planned that I plan on just sinking my teeth right into a backhoe. That's usually how I do it, but. But yeah, no, I, you know, I'm an avid aquarist, as they say, so I've got a couple of reef tanks at home. [00:14:12] Reggie: That's really cool. [00:14:13] Kane: You find ways of continuing that path too. And I live. My brother, who's also a UVA grad, went on to do molecular cell. He's a cancer cell biologist and so he went on to get his PhD. So I feel like, you know, at least one of our heads went in that direction. We're at least fulfilling kind of, you know, a societal like, I want to give back right from the gate. [00:14:33] Reggie: I love it. [00:14:33] Kane: I'm the sellout. [00:14:35] Reggie: We wouldn't call it that. [00:14:37] Kane: No, no, no, no. [00:14:38] Reggie: I mean that I wanted to ask you all too, is kind of about this emergence into the field of data science. It wasn't necessarily scoped when you all were studying and in school and beginning in your careers and things like that. And one of the reasons why I think it's interesting is because we're. Society is talking a lot about AI and things like that and kind of what the future is, but it's not quite scoped as well. And so you all have this experience of stepping into an emerging world where you could see some things forming, but it wasn't quite concrete. And so I'd love to hear when you first realized that your work that you were gravitating toward was becoming kind of its own thing, a real discipline that was legible to other people. And how have you seen it change since then? [00:15:22] Heidi: I'll start. So it's really interesting. So I'll go back to my first job out of college at that unique group at PricewaterhouseCoopers. And our partner in charge was a PhD statistician from Stanford. And I will never forget it was probably my second year there. So I'm 23 years old. This is 1992. Okay, so think about. Most of the people listening probably weren't even born. Then he said, In 10 years, all this work that we're doing, coding in SaaS, we use SAS a lot. It's all going to be automated, and companies are not going to need us because they're going to be in control of their data and they're going to use it to make decisions all day long. And we've got this Runway of about 10 years. So that's 1992 to 2002. Look where we are. It's 2026. [00:16:20] Reggie: Well, it's because Y2K happened. [00:16:22] Heidi: Exactly. So now I'm not saying that the. Let's just sit back, everybody's safe, it's copacetic, and we don't need to be really thinking seriously about how AI is augmenting humans in the workplace or taking over jobs. But my prognostication, I guess, is that this is going to be a ramp. And we hear all these great stories in the news and on LinkedIn and we are interacting with this technology day to day. But the vast majority of companies that I work with and consult with and have as clients, they're not the top 5%. They are still struggling, frankly, with the same things that I saw in 1992 when I was a entry level consultant working at the Limited or Harley Davidson or United Airlines. They're still struggling with those same things. And I think a lot of companies still have problems with accessing their data and people consuming it are still struggling a bit with their data literacy and AI literacy, meaning knowing how to use that to make the best informed decisions. That linkage part is still hard for a lot of people and businesses. [00:17:53] Reggie: Yeah, that makes sense. [00:17:56] Kane: I mean, I guess my narrative would almost be fully complementary to that because I started out working under a PhD, Michael, a PhD in econometrics. And again, that just kind of furthered my passion for. I want to, you know what? I want to be a lot like him, but then coming to the realization that I am not as smart as him. And so what we became really good at in the beginning was, I think, the realization that we're really good partners. So I think with an analytical mindset, I was able to actually convert the decision scientists. So Heidi would be on the other end kind of barking down, hey, there's a problem here. I can't get at this. I can't do this. And So I, for 15, 20 years was either on the front line working with clients to build a fully enabling data foundation from policy, standards, governance, quality, creating ownership and stewardship around it and making sure that the ones and zeros that we were serving up to my clients, the, the analytics coes at clients or the big rock program drivers at clients or my internal stakeholders, which half of our PhDs in a data science type of realm are telling me, I need this, I don't want to be doing this. This should already be taken care of. And so 20 years I've been building that foundation and same thing, you're like, Project 1995, we've got to go back to this to be like, yeah, our analytics people are actually spending 50% of their time cleansing the data or finding areas where it's wrong. I wanted to solve that. So this internal role has kind of given me the remit to at least start doing it. It's a big world out there. [00:19:44] Heidi: I think that the net net is, these are cycles, these are technology inflection points and there's a ramp and there's the top 5% and then there's everybody else. And certainly, the ones at the bottom that don't learn how to evolve and adopt are going to be not around much longer. But there's this whole big opportunity in the middle. And I think we've got a lot to work with for the foreseeable future. So I'm optimistic about what we can do in our fields with AI and data science. [00:20:20] Kane: I am too. And it's funny and it might be a little cynical. I don't know if this will. But after every major hype cycle, there's always a lot of, you know, there is that 5, 10, maybe even 20% that comes out. That's it's, it's business scalable, it's efficient, it drove the ROI like great. Then there's all this other stuff that you know, people tried, you know, like little bits and pieces. And I think that, you know, what we have done is tried to create a discipline around like let's align ourselves to a couple of, like the, the big rocks, like a couple of things that we can all get behind. Because when the dust settles, there's a lot of cleanup and there's kind of this constant cycle of doing that to be like, man, we got data everywhere. Wait, when we did that, we put data everywhere over here. We got to go clean that up now. Now we've got, you know, at one time it was citizen led chatbots, you know, everywhere. We got to go clean those up because some of them are clobbering each other. Same thing is going to happen probably with agents without a clear kind of governance framework around it. And so we Worked very hard to get in front of that right now with a business led kind of approach to it. But I mean, I agree it's kind of almost like an evolving process. Go out and build, figure out what worked, clean up a little bit and kind of just progress the cycle. I think those will compress. [00:21:40] Reggie: Yeah. [00:21:42] Kane: But yeah, Heidi, I mean, I agree [00:21:45] Reggie: it's interesting that you're talking about like having to clean up some stuff afterwards. I don't know why brains work in interesting ways, but before we started recording, Heidi, you mentioned something that I wrote down and asked if I could come back around to where you said I hate the term quick wins and low hanging fruit. And it just was this. And I'm thinking about cleaning up the low hanging fruit because of the way that you framed your relationship to that phrase. But can you say a little bit about why you don't like those things that are common colloquialisms in business? So it seems like a lot of people like them, but you have a different perspective. [00:22:20] Heidi: Yeah. So I mean, and I guess this will. My vantage point is probably similar to Kane's and I come in as sort of a consultant. But if you think about, you know, let's say an organization is like we need an AI strategy or we need to get better at making decisions using our data and AI. And so what typically happens is they'll probably try it internally and they go for quick wins. Or the other term that I just love is work with the willing. So that means people who get it and you don't have to like beat them into submission to understand like you should be making decisions using data or you can totally AI ify this process. You don't need 600 people typing in financial statements or whatever. And so you work with the willingness, or you get a quick win, which means it doesn't take much time and it doesn't cost a lot of money, but it's got like a big dollar. Right. And so what you've essentially done is you've limited your decision making on two variables essentially or two factors, which is return on investment or how hard it is and life. And businesses are not that simple. They are multi-dimensional and they require so much more than just top line cost savings or revenue additions and how long it takes to get it done. There is the people and human component. There is the how hard is it to get Joe and Sally to change how they do their day job, which is the big one that we don't spend enough time talking about. There's is this tied to Strategy. And I was sharing earlier that I was working with a client recently that had a mandate to AI ify an entire operations organization. And so they focused on working with the willing and getting a quick win. And that quick win drove a 40-man hour per year savings. And that was a womp. Womp. It didn't. It was quick. It was quick and it did. And by the way, that particular thing that they aified, if that's a word now it is now wasn't even tied to one of the product lines. That was a growth strategic product line for the company. And so you could argue we shouldn't be, you know, we shouldn't even be focusing on something like that. And that, and that's why I get frustrated with the term quick win or low hanging fruit because I feel like it's also maybe taking the easy path and it's not looking at the forest for the trees where you see all the nuances of things and life is complex. Businesses are complex, processes are complex, people are complex. [00:25:22] Reggie: Absolutely. I'd love to hear a little bit, Heidi, about your path to entrepreneurship with NavAlytix AI and how you decided to start doing that and tell us a little bit about it. I mean, what's the kind of premise of your company now? [00:25:39] Heidi: So how I got there, I will say, as Kane mentioned earlier, I mean it's scary to put yourself in a very uncomfortable position where you've never done something like this before. I mean I'm used to getting, I'm [00:25:54] Kane: in awe of it too because I feel like I kind of like I took the easy path and Heidi's going to talk about like taking a big leap of faith and taking a risk. And that's probably in like half of the students minds that we're talking about is like how do I do that? And I think telling the story is pretty awesome. [00:26:12] Heidi: I mean it's risky. So I will say same thing. Kane mentioned the need to put food on the table. I started this right as my youngest was in the last year of college. So for me that financial stressor was going away and fortunately my husband has a good full-time job that has benefits and all of that. And so he was fully supportive of this. And what drove the decision though was an incredible amount of opportunity as well as frustration with this thing that I see happening over and over and over again. And I've seen it since I was, you know, 22 in my first job. And that is there's so much potential for companies to make decisions based on data or automate something using Data with agentic AI and it just struggles to be sticky. And as a former chief data and analytics officer in a couple of companies, I feel like I did all of the things. I talked to lots of people. I aligned my strategy to the company's strategy. I made some mistakes by doing some quick wins. Absolutely. And I learned from that. But I just see, there's this. It's the nut I wanted to crack. I mean, I feel that this is achievable and I really want to help other people. Folks in the Chief Data and Analytics, or Chief Data and AI Officer, whatever we're calling him, someone in charge of data and AI strategy and transformation. I believe that there is a better way to do that. And no knock-on consultants. Kane. [00:28:10] Kane: Here's where the podcast goes sideways. [00:28:15] Heidi: What we do is we basically take a three month lots of PowerPoint decks, lots of workshops, lots of whiteboarding sessions, more PowerPoint decks, lots of Excel sheets on use cases, and lots of big expensive bills from consultants. We do it in three weeks. We do use some AI to do it, but our company's been building a trove of industry specific use cases and we tie those benchmarks in with a company's financials and their strategy documents and their go to market plans and their product roadmaps. And we basically have AI fied and automated the heck out of a consulting engagement. And coming from a consulting background, I would call it 75%. It's a platform and it sits with companies and it's their AI navigator is what we call it. And it's their roadmap and they upload their successes as well as their failures and then it rejiggers the roadmap and that roadmap is changing and it measures roi and we put a little light consulting wrapper around that. But it's mostly a product and a platform and it's the nut that I wanted to help people crack and that's why I'm doing it. [00:29:35] Reggie: That's incredible. And you're making me wonder now as you're building this company in the age of AI and agentic software and vibe coding and all that. And we can talk about vibe coding if you want, but I'm thinking about how did you staff up to build out that platform? Do you have full time folks? Are there fractional folks? Did you hire consultants? Are you doing all of this yourself? [00:30:00] Heidi: Vibe coding? [00:30:01] Reggie: Yeah. Are you vibe coding it up? [00:30:04] Heidi: My partner does some vibe coding and then we've hired contractors to actually make it ready for prime time and get all the past the infosec and security hurdles and things like that. So that's the piece we're working on now. It's basically two and a half of us and we're risk takers. We've obviously quit our full time jobs and, and we are bootstrapping this ourselves with our savings accounts and not taking paychecks for a while and trying to land some good codev clients which we have had two and we're fingers crossed about to sign two more. [00:30:55] Reggie: That's great. Can you define codev for the audience? [00:30:58] Heidi: Yeah, sorry. So co dev is short for co-development and it's basically you're looking for a client and when I say client I mean a paid client. And so one of the things that I have learned in accelerator that I was just recently in with one of Kane's competitors, so we were in an accelerator that was for tech startups that we got chosen for, based in Chapel Hill, North Carolina. And they bring in experts who are just great, awesome in the startup community and they say don't do anything for free. You might not, you're not charging, you know, you're used to be consulting rates, but if you don't charge something for what you're doing, people won't show up to your meetings, they won't test your product, they won't give you feedback and so you need the skin in the game like Kane said. And so they, there's a gift to get. So they're getting a discounted price and you're getting an incredible amount of feedback and learning and they're getting to influence your product roadmap and they're not paying much for it. Yeah, that's what the term is. [00:32:11] Reggie: That's great. That's super helpful. Yeah, that was helpful to understand because I think that these are those little signals that kind of give us some ideas of what the future of growth and future of companies and building and hiring and things like that look like. One of the things that I wanted to ask about was some stories. I mean people hear data science, they hear analytics, et cetera, and maybe have some loose ideas. I mean even still people cite Moneyball as a classic example and there are so many other examples that have existed throughout you all's careers. Can you all give us a couple of projects or stories from the type of work that you've done to really kind of illuminate what data science has looked like for you all? [00:32:58] Heidi: Okay, well, I feel like I've been doing a lot of talking. [00:33:01] Kane: No, that's all right. [00:33:04] Heidi: I got stories. So we want to talk about the bluefin tuna population. Harley Davidson, motorcycles. [00:33:11] Reggie: Let's do tuna. [00:33:12] Heidi: Because we're going to talk about tuna. [00:33:13] Reggie: Tinned fish is hot right now. [00:33:15] Heidi: Okay. [00:33:15] Kane: Also, I mean, I'm going tuna fishing this summer, so I'm. [00:33:18] Heidi: Okay, okay, so this will be relevant. [00:33:20] Kane: There we go. [00:33:20] Heidi: This will be relevant for my partner here. [00:33:23] Reggie: Yes. [00:33:24] Heidi: Okay. So bluefin tuna. So for those of you who are anglers or interested in recreational marine fishing, the bluefin tuna population is actually regulated. Oh wow. On the coasts, Pacific and Atlantic oceans. And that is because we don't want that particular species to be overfished. And so one of the things that I did in one of my summer jobs in between years at UVA was help with surveys of recreational and charter boat captains. And they would survey how much fish were being caught as people were getting off of charter boats and things like that all up and down the east and west coast. And that data came back in and was used to estimate based on seasonality, the ocean temperature and how much was being fished, how much was being thrown back alive or dead, how much was being used for bait. See, this is right up Kane's alley. He's like, he's into this. [00:34:29] Reggie: Exactly. The analysis. [00:34:30] Heidi: Yeah. [00:34:30] Kane: And so it blends the ecology and then my trips this summer. [00:34:35] Heidi: Yeah. So the data that we collected and then the forecasting models that we produced would actually be used to put the signage that would go at places where people were fishing and issues to charter boat captains on. Yeah, you have to throw everything back that's over X number of pounds. And so that's an example of something where. I mean, I will say I was not exactly enthused about counting fish. I wasn't the one out there doing the interviews, but I did get pieces of paperback that had like fish guts all over that. And we would have, we would have people doing. This is pre scanning. So we would have people doing data entry. And then I was part of building the models that were being done. [00:35:17] Reggie: Yeah, but true data cleaning. [00:35:19] Heidi: But. Yeah, but then you, but you really got to see. And I think this was the thing that was impressive to me, even though the use case was not. I'm sorry, I didn't really enjoy fishing, but. And I still don't. And I don't know why they call it a sport. Like, do you believe that it's a sport? [00:35:34] Kane: I gotta tell you, if you are on the other end of the line of a 40 pound tuna, you're, you're burning calories. And if it's a 250. If it's a bus, you're not gonna have to work out for a month. [00:35:46] Reggie: Okay. Just my perspective on you is what it takes. [00:35:50] Heidi: Okay. Anyway, we digress, but it was really neat to see that that was actually helping someone or something or the fish in this case. So that's an old example, but we have lots of examples. What about you? [00:36:10] Kane: I'll go back to my early days, which really kind of invigorated my passion for blending data and analytics and advanced analytics with technology. And it was a lot of early work in the telecom industry around different product uptake and being able to blend the marketing and the personalization area. Because at the time it was massive mass marketing approaches, flyers in the mail. And then all of a sudden everything began digit, you know, was digitized like [00:36:44] Reggie: my AOL CDs back in the day. [00:36:46] Kane: There you go. Yep, I remember those. You know, with everything you got, it felt like you got an AOL startup. [00:36:53] Reggie: Yeah, exactly. [00:36:55] Kane: But yeah, I know, I mean analyzing though, you know, the take rates and like, you know, walking, helping companies kind of walk through the, the buy flow where people are coming out. Honestly, the data collection aspects of that is probably where I made more of a kind of a brand for myself, at least within my teams. But when you put it all together and to be able to move from mass to micro segmentation to personalization over a 5 to 10 year kind of timeframe was a pretty marvelous. It really showed the power of collecting data, doing some curation and cleansing data and then putting it in the right minds with the context of the business to ask certain questions like is this segment going to do this? I mean, you know, back in the day, you know, you're sitting there going, there's no way people are going to actually look at their 4 by 6 iPhone screen. Right? I don't even know if there was an iPhone at the time, you know, Android. How are you going to get this on a BlackBerry, you know, and watch a movie? That's stupid, you know, that's crazy. And then we started testing it, you know, with some of our partners and like you're like going, oh my God, this might actually take off. There are data products that we're going to create. You saw the entire evolution over a decade within the telecom space leveraging the power of data and analytics. It rode the evolution of just voice only all the way into a data product and then just full blown video streaming and then the adjacent spaces that we would play in. In terms of where are you going to buy Spectrum and Are you going to be in MVNO or an MVNE and actually create these virtual networks and you're going to try and supplant one of the big providers? That was a really interesting time. And the power of data. It was just that proliferation of data variety, volume, the 3vs, 5v, 7vs, whatever, that was a really cool time. And honestly, taking the outcome when you build some of these things. And then, you know, at one point I was at a client and we were working and it was 85 degrees in the chief Marketing officer's office because we had a server that we had brought in shadow id, you know, because again, like you, sometimes, you know, your risk takers have to be cowboys or cowgirls to do some of these things. [00:39:18] Reggie: Be cavaliers. [00:39:19] Kane: Yeah. Little cavalier about it, you know, like be mindful, you know, collect all the, this data, but then traverse the world into. How do you actually scale this into customer behavioral economics and choice optimization? And how do you present this through kind of a web space and actually truly scale a really defining data analytic asset? That was awesome. [00:39:44] Heidi: So don't you think, though, that AI now actually takes those things that, that we've been building for the past couple of decades in our careers and makes it more accessible and more easily adoptable in some cases? [00:40:02] Kane: 100%? I think that that is definitely the premise. But when you look at that, and I think it's totality, I think it's the next concentric circles that are going to create space for a lot of us who are thinking about the core. And then, all right, well, what if. Where does the knowledge and the wisdom go through that process of creating that? And it does, it democratizes the analytic. But then there are kind of adjacent spaces. Where does the decisioning process actually land? And is that actually going to be integral to the next best decision? And how does that get layered in? Who owns that? Actually, I mean, I'm working in multiple geographies where this is actually a serious question. I created an intermediate of data coming from. It's got some, forgive me, but it's got some US DNA to it, some data, and it's got a little bit of German DNA and it's got maybe even some Southeast Asia DNA in this great thing that we just created in this analytic. And it's like from a governance perspective, who owns it? And is it obfuscated enough to say no, it's actually in a different domain and we've got to kind of redefine an ownership model around it. These are not the, you know, the, the AI driving, kind of headline driving things. But these are the foundations that somebody's going to ask five years from now to be like, wait a minute, where is all this going? Yeah, I don't care if it's in a vectorized data. I can still, you know, I can still get it out. We can reverse engineer it. [00:41:33] Reggie: Especially with Mythos, apparently. [00:41:35] Kane: Well, and so, yeah, I mean, you know, I'm in the foundations right now, but you got to be constantly thinking ahead. But your point is, right? I mean, a lot of these things are just going to make it faster, better, cheaper to get to a yes or a no answer. And then, you know, to your other point, I do not like low hanging fruit or, you know, I mean, that rots. Typically, [00:41:59] Heidi: it's a bad term. [00:42:01] Kane: Well, but I think these are the ways you get to. You don't have to deal with that actually anymore. You get to those answers faster. So you can either fill or kill something pretty quickly. [00:42:11] Reggie: Yeah. I mean, I'm curious though, because this does. When I hear speed. I mean, we mentioned vibe coding and passing earlier and joking, but we did talk about it a little bit earlier this week and we were kind of getting to a point where we were talking about expertise and the difference between kind of being surface level and going a little bit deeper. And Heidi, you shared an analogy that resonated with me as a foodie and as someone who's in a cookbook club. And so I'd love for you to maybe share a little bit about your thoughts on that. [00:42:40] Heidi: Yeah, my chef analogy. And I think we were talking about the term that was, I think prevalent, I don't know, five to seven years ago, the citizen data scientist. And now we talk about vibe coding. And I think they maybe misrepresent the deep expertise that's needed in any of those disciplines. Data science or coding. Vibe coding is great for building a really cool prototype to get your art of the possible out of your brain and into something that maybe an engineer could then look at and then go build it in a way that's close to what you're thinking about. Citizen data scientist. That scared me a little bit with the dangers in building predictive models that actually could be life impacting to people and to businesses. Like, I've seen some examples of companies getting either missed over for like small business loans or given small business loans when they were not scored properly because citizen data scientists in the marketing department's getting pressure from the North American sales leader that we need more accounts. And so I'm going to go Build this model and then we score people and then they get through and they shouldn't. Stuff like that can happen. And that's similar to what people I think are a little afraid of with AI. But the chef analogy, this is where I think my advice to students, grad students, undergrad students, is really try to go develop a specialty and some deepness in your skills, whether they be in data science, whether they be in the human side of things, whatever it might be. This is really applicable to any profession, I think. And so if you think about a chef that is hired to go in and remodel a restaurant and they're given some staff and they're given the tools and they're given the pantry with the ingredients, if they really know their art and their craft well, they are going to be able to say, I need a different stove or I need a different kind of oven, or I need these other tools, or I need people that have some different skills, or I need some interesting fresh herbs and things like that. [00:45:11] Reggie: I need bluefin. [00:45:12] Heidi: I don't need skipjack. Exactly. And you know why? Because you spent years experimenting with the science of how ingredients complement each other. And I think if you are not a trained chef, so we'll use our citizen data science analogy to, you know, just a citizen chef. You don't have that background and that expertise and that training. That's not to say you shouldn't go cook for your friends and your family. But when you take a profession seriously, when you bridge from hobby to profession that affects lives of other people and other businesses, I think that depth and credibility is absolutely necessary. [00:45:59] Reggie: Yeah, yeah, it definitely seems like that's what we're seeing in the market in general, where there's this kind of shift from credentials based hiring and recruiting to skills based hiring and recruiting. And so now there's a lot more technical assessments and take homes even on the front end, before you pass the first round. And so it's, it's interesting to hear that folks are also looking for GitHub repos and portfolios before they're looking for resumes and things like that. We're just seeing that become a trend more and more. To your point, I think I wanted to ask one last question to kind of close this out. I won't ask a big AI question, but I will ask an advice question, as both of you all are alums, if you could send one message back to yourself as a UVA student, not about technical skills, but about how to navigate being new and being new in an uncertain world. That's moving fast. What might that be? So, like, if you put yourself back in your student shoes and you knew that you were about to enter this big world, but you have the hindsight of 20 20, what advice might you give yourself? [00:47:09] Kane: And we have the reach of digital. So if that were the place, I would say again, I go back to inflection points for me, and when I realized that how important mentors were, and this was upon probably 15 years of reflection, when a mentor would leave or kind of move on, the gap that it actually created is kind of what created my inflection point to analyze. Like, wow, that felt like I just went into a void for a year or two, you know, networking. And I said it to our students today. I say our students, your students, but these students. [00:47:46] Reggie: You're part of the family. [00:47:47] Kane: No, I know, but. But to take an opportunity and take a little bit of a leap of faith to actually reach out to somebody and have a conversation. A couple of folks always after these, will come up and have one on one. And it's so good, and it's good to see. You can't change everybody. You can't put everybody on the right trajectory. But if you have an incoming funnel of people and you take the risk of approaching someone, I wish I had done that. I wish I had asked more pointed questions or I greet everything with curiosity, not ferocity or fear. And I don't know, I think I would have probably accelerated my path to a couple of good outcomes as opposed to flail around a little bit. But that would be the. I mean, there are so many things I would be giving my younger self advice on, but that would be the big one. [00:48:37] Reggie: That's a great one. And I love your curiosity over ferocity. [00:48:41] Kane: I got that from Paul Griggs and Mohammed Gandhi. I sold Mohammed, actually say it coming out of Davos. And I think it's just absolutely kind of an amazing thing. I tell my teams that now all the time, because we're in an environment now where you can't have a meaningful conversation about a controversial topic, let alone you're debating whether or not you can actually, how do we go and build this? And everybody's got their different perspectives. And so from top down, bottom up, it's like, ask why. Why do you think that way? Why do you think that's a better way to do it than that? And I feel like there's. It opens the door to a faster, better, I think, more agreeable outcome. [00:49:22] Reggie: And then, Heidi, how about for yourself? What advice would you give Yourself, probably [00:49:26] Heidi: similar to Kane, I would say I'm going to go back to one that he mentioned earlier, which was about taking risks and growth and feeling like you were really growing when you were in the most uncomfortable positions ever in your career. And I would suggest to seek those out. Like, seek them out. Don't wait for them to come to you. One of my most growth, biggest growth cycles in my career was when I was leading global data and analytics for Nortel, which is a bankrupt Canadian telecommunications company. And, and a year into my tenure there, they filed chapter 11. And I mean, it was like every quarter, 30% of the people were getting laid off and I was reporting into the chief marketing office and then roles got consolidated and that leader got to run sales operations and asked me to actually run sales operations for North America. And I was like, I've never run sales op. I've been like this data and analytics person. But thinking about it, it was like, wow, this is my opportunity to make people actually listen to all the data and analytics that we're doing. And I got to influence how we were spending our money on programs. And then I got marketing and marketing operations and the marketing campaigns team. And man, it was uncomfortable because everybody on that team knew that I did not come from the same trajectory that they did. And so. And it was incredible amount of hours, more hours than working for PwC, honestly, on a weekly basis. But I learned a ton. Like, I learned so much in like a year and a half. That's incredible. The other thing I would say is break bread and talk to people who do not look like you. And I don't mean on the face, I don't mean color of skin or gender or anything like that. I mean like the kind of work that they do. And you will learn a ton, especially if you ask them what's really hard about your job. And you think about that with your data science hat on. And it brings me back to a project that I was on for a large clothing retailer and I was sitting with my, like my cubicle was actually with all of these fashion merchandisers. They did not know and understand math, like at all. And they used calculators and they wrote numbers down. They weren't even using Excel and just sitting with them. Well, first of all, it was really fascinating to just. I loved experiencing what they were experiencing because it was so different from my job. But I got to see all of these opportunity areas for where they could have data be like an incredible enabler and catalyst for them. And so asking them questions and being curious about people's jobs and roles that are completely different from yours. You will grow leaps and bounds from that. [00:52:48] Reggie: I love that you all both have curiosity as a core component. [00:52:51] Kane: You know, when you actually just to extend and I agree like that curiosity and kind of opening your, kind of your brain up to like just a diverse perspective. Right. I actually think, you know, we have, at least at PwC and I think just generally speaking is that we see, you know, we say AI is lifting the floor and humans are gonna lift the ceiling. Right. And I think that like when you actually embrace it and I think the real, the people who are going to succeed kind of of embrace it and actually bridge that together to say I'm going to seek, I'm going to seek perspectives from my agents, my AI enablement layer. I'm going to bounce it off the humans that I respect and I follow and bridge those two things together. I think there will be just an absolute continuous gap up of productivity and impact on whether it's a client project or society in general. I think that that's the way we're going to ladder up. [00:53:50] Reggie: Yeah, I love it. Ladder up. That's the conclusion. That's amazing. Thank you both for your time. This was incredible. [00:53:56] Heidi: Thank you, Reggie. [00:53:57] Kane: No, thank you, Reggie. Appreciate it. [00:53:58] Reggie: Absolutely. [00:54:03] Margaux: Thanks for listening to this episode of UVA Data Points. More information can be found at datascience virginia.edu. if you're enjoying UVA Data Points, be sure to give us a rating and review wherever you listen to podcasts. Podcasts. We'll be back soon with another conversation about the ever evolving world of data science.