Demystifying Data Science with Dr. Hisham Talukder

Unlock the secrets of data science's transformative role in business with our guest expert, Dr. Hisham Talukder. In this episode, we dissect how data science has evolved from the ivory tower to the tech trenches, where statistical acumen and coding chops reign supreme. We're untangling executive misunderstandings and advocating for the strategic deployment of data science—not as a panacea, but as an essential precision tool aligning with organizational ambitions.

Prepare to navigate the captivating tides of generative AI and its disruptive potential across various sectors, especially the game-changing ad tech landscape. We're spotlighting the importance of thoughtful adoption and incremental project launches to ensure congruence with corporate visions and to unlock the transformative power of this technology. Tune in to sharpen your foresight and join the conversation at the intersection of data science innovation and business acumen.

In this episode you'll learn the following:
1. Understanding that data science is a tool that needs to be used appropriately and strategically.
2. Importance of interweaving data-driven approaches with established business frameworks.
3. The role of data in decision-making and problem-solving.

CONNECT WITH HISHAM:
https://www.linkedin.com/in/hisham-talukder-b20b6647/

This episode is sponsored by LucidPoint
Are you struggling to take your IT organization to the next level?
We help our customers do so with confidence. Turn your vision into reality, call LucidPoint today!
https://www.lucidpoint.io/

CONNECT WITH SUSIE:
https://www.linkedin.com/in/susietomenchok/

CONNECT WITH JAMES:
https://www.linkedin.com/in/capps/

Transcript

Speaker 1: 0:15

Welcome to the Quick Take podcast, the show where you get targeted advice and coaching for executives by executives. I'm Suzy Tomlinchuck.

Speaker 2: 0:24

I'm James Capps. Give us 15 minutes and we'll give you three secrets to address the complex topic of issues that are challenging executives like you today.

Speaker 1: 0:35

Hey, welcome to Quick Take. I'm your host, suzy Tomlinchuck, along with James Capps. How are you, james?

Speaker 2: 0:40

I'm proud to be here.

Speaker 1: 0:43

I am super excited too, because you brought a friend along. You brought a friend.

Speaker 2: 0:49

I brought a friend along. It's like he's my chaperone but, yeah, very excited, we've got a good friend of mine, dr Hitchin Tolerctor, here today who has got a great background in data science, super sharp individual I've learned a lot from in this space. We thought today would be a great opportunity for us to talk a little bit about data science, talk about the history and the evolution of that space. Gosh, so much is going on there, so many of our guests and our listeners are asking questions about that. We thought we'd bring Hitchinman to talk a little bit about that.

Speaker 1: 1:31

Dr, this is a person that knows what he's talking about. I'm in awe. Thank you for bringing him. Yes, yes, I have a new friend too.

Speaker 2: 1:40

There you go, Hitchin. Why don't you just set the stage for us on this ecosystem?

Speaker 3: 1:46

Yeah sure, James, Suzy. Great to be here, James. We're officially friends. I have it on video now so you can deny it from my friend.

Speaker 3: 1:55

I think my experience in data science is very unique because I've seen the evolution of data science. I've been in this data science field for about 10 years now. I come from a PhD in applied statistics. I remember my first job out of grad school it was like I was only working with PhDs, so it was a team of four to five data scientists, all PhDs, Two.

Speaker 3: 2:17

If you look at the data science field now, hey, if you know how to code, if you know statistics, if you know math and you know how to code really, really well, you can be a data scientist. And that's not to say that the skill set has gotten down. It's really around the evolution of technology, the evolution of computers, that have seen that change. And I've worked on data science teams where you're in the back room all you're doing is coding, and I've worked with data science teams where you're working with the most senior leader in an org or a company to try to understand problems that you could potentially solve with data science and any spectrum in between, and obviously you need different skill sets to tackle all of these different problems.

Speaker 1: 3:05

Yeah, and I was thinking when you were talking about that is such a different place than I am. I am not a technologist and so I would never be in those rooms. And I actually was just meeting with a friend of mine who's a C level person and she said I feel a lot of pressure to know more than I do and it makes me really nervous, and so I really would like us to really unpack that. What do you see at those high levels about what their assumptions are?

Speaker 3: 3:33

Yeah, I think one of the biggest things that I've seen consistently is this data science being treated as a silver bullet, a black box. You give me a problem, data science will solve it. And there's always this discussion, whenever I join a company, whenever I join a new team or even an existing team that's been there for a couple of years, of talking to leadership and trying to explain to them what data science is. Data science, like any other role, like any other field, it's sort of building a house, right? Like you need to first build a foundation, you need to build on top of it, you need to start with the easy problems. And then one other thing is data science always needs to match the business that it's trying to help. Not all businesses need Gen AI, not all businesses need this complicated solution, right. So it's like really working with leaders, not as a way of educating them, but almost partnering with them, to say, like, here's what data science is, here's what data science is not. Now, how can I help your business in the most effective way?

Speaker 1: 4:40

I think also, what you're saying is don't be afraid to ask the questions like how does this fit together? Because to me, from a person that is, I was in the business piece of it, when I'm working with somebody that really has an expertise around data and technology, to really just be like how do we make this work?

Speaker 3: 4:59

Yeah.

Speaker 1: 5:01

Like to your point about the baseline. How do we work best together?

Speaker 3: 5:04

I think that's a really good point. It's not just at a human level how to work with partners but also at a technology level. Where I have seen in my experience data science be the most successful is when it becomes a part of the overall system. For example, I worked in an ad tech company where we wanted to make sure we're showing the ads that is most relevant to our end user.

Speaker 3: 5:32

If you think about that system of showing that ad people come into a website. It fires over requests, ads get sent, people click it, then they go and may or may not purchase it. That's the system. What you don't want to do is come in and say, hey, let me do an analysis and let me tell you build a PowerPoint slide of what's going on. You want to make sure you build a solution that you could almost insert into a system. Now the system becomes people come into a website, they request an ad. The ad goes to my data science system, gets the scores out, it tells it what ad to show, then it goes back and continues it. It's not only just building a partnership of togetherness at a human level, of making sure you have the right partners and you're working together, but also at a technology level, and making sure everything is working together and talking to each other. Wow.

Speaker 2: 6:24

I really see the need to decompose that mystery. Really. Take from hey, we've got this new thing I saw on the InFlight Magazine and I want my company to implement it to hey, let's take a look at what we're trying to solve, for what are the problems we have and how do we articulate them? I love that jump from magic bullet to individual problem solving. I think that's something that so many leaders lose track of and expect. Maybe my data sciences will just solve my problem.

Speaker 3: 6:58

I think where that is the most successful is, once again, when it's a partnership and you're working together. I always tell my business partners you know the business a lot better than I do. What you don't want to do is come to take lessons from me of here's a business problem you should be solving. You should be bringing up a business problem. We should have a discussion, and I know the technology, you know the business. That's where we make the best partnership. It's not a one way street. It's not them telling us or us telling them. It's really working together. And I do think that evolves as you gain more and more trust with your partners. As you gain more and more trust, sort of the opportunity start to open up. Right Now we can take some big bets that they might not be thinking about because they have that trust that, hey, we've been able to build certain things. That's become ROI positive.

Speaker 2: 7:52

So what I'm hearing you say, if the first real you know the nugget we want to talk about, is really realize that it's not a silver bullet the second thing I'm hearing you say, though, is you've got to create that trust and you've got to create that dialogue, and to do that, you really need to start small, and is that really the best way you've seen people?

Speaker 3: 8:10

work together. That's right. I think you need to start small and you also need to, going back to that house analogy, build the foundation and sort of a vision of how that will evolve, right? So it's like, hey, for example not surprising to anyone data science needs data. Data needs to be part of that foundation, right?

Speaker 3: 8:33

Many times I've been in sort of the final presentation of a data science model where they will bring up issues, but it's not really issues around the model, it's issues on the data, right. And it's like hey, like yes, but we've assumed that we have issues with the data, right, you need to assume certain things about the data to move further, because what you can do, also on the other side, is wait for perfect data, right? So, like, you need a foundation of data, you need a culture of data. The culture of data means, hey, data science not all the time will work. You need that ability to fail. You need that ability to learn from your failures. You need the ability to experiment. So I think, like, If you pair up data, if you pair up the right culture and the right buy-in from the senior leadership, that's where you start to evolve this culture of data science. You can put a vision forward that is not just project to project but more around business challenges to business challenges.

Speaker 2: 9:39

Yeah, I think that it's so fascinating how so many of our implementations, our journeys, our changes are really foundational on culture. You can do a lot of things. You can bring in consultants, you can bring in new tools, technology. At the end of the day, you've got to have a conversation around culture because, as I always say, the technology is the easy part, the people is the hard part. That culture has got to be part of the conversation. One thing around.

Speaker 3: 10:10

That is, that culture doesn't just stop. You don't need investing in that. Culture doesn't just stop. I've worked with the most technology savvy company to startups that are just getting into technology Both of those sides. You need to continuously invest in the culture. You need to make sure the culture is there, but also you need to make sure culture is evolving. The culture that might have worked when your company was 10 people with one data scientist is probably not going to work when it's 100,000 people with 2,000 data scientists.

Speaker 1: 10:50

Especially at the top in leadership, because you can assume that everybody has the same mindset as you and, as things are growing, that everybody is on board and energetic. It is really important. I love what you said about the relationship piece and that trust, re-evaluating that and re-syncing around making sure everybody understands the parameters, if you will, around each of the business challenges.

Speaker 3: 11:19

Absolutely. I always told the story to friends. I always realized I made it in. A job is when I have a senior person that's not from the technology side, either the business or product side just calling me out of the blue and saying hey, I just want to brainstorm something with you.

Speaker 3: 11:37

That's when you know, as a technologist, you've made it in that room. It's not just a technology thing, it's like, hey, I just want to brainstorm something with you that might start at the high level. There's definitely a data science element to it and we're going to get to that. That happens to me at every job. That's when I'm in my head I'm like all right, I've made it, I've arrived. I think I've made it in this job. It's a great feeling because it goes both ways. You've shown the product and business leader that there's value you can bring, and also they see the value in a technologist's leader. They see the value in a data science leader.

Speaker 1: 12:17

It makes me think. As a salesperson, I was resentful of the engineers, the technology people, even the finance people.

Speaker 2: 12:26

Wait a minute. Why ask me all?

Speaker 1: 12:27

these questions.

Speaker 2: 12:28

Wait, do you not resentful me? You love me, I know I was like just get out of here.

Speaker 1: 12:35

I'm going to tell you when I need you. You're right that influence is so important. One characteristic of influence is when you get that call. People depend on you. I didn't expect this conversation to go to relationships and culture, but it really is great to know that both are so important to the success of this kind of project.

Speaker 3: 12:59

I think you learn it with experience, because I'm sure there are people that I work with in the sales department early in my career that would agree with you, Suzy.

Speaker 2: 13:08

It's like hey, I hate to work with that guy at shop.

Speaker 3: 13:11

He thought he knew everything. But then you kind of evolve, and that's where I meant of just building that partnership I think is honestly one of the key factors of data science success. You need to make sure you have partners that you could rely on, that you know you could work with to ultimately solve the problem.

Speaker 2: 13:30

Yeah, it's, you know it's always about the people. We say that so often on the show, but and I do think this is a unique space you know there's so many technologies that are that are people adjacent or people use, but this is problem. This is looking at problems and addressing issues that are very complex and are not simply just an implementation of a piece of hardware or a new plant or some code that's, you know, doing additional information security.

Speaker 3: 13:54

This has the opportunity to really bridge a lot of areas that I think you're you're highlighting a really unique part of this, of this emerging space that I think we need, we need to take into consideration and and the space Continuously evolved right, like, if you think about once again going back to sort of my journey, when I came in like statistical models, like that was data science, right, and then it went machine learning. That became sort of a big, big, sexy word in data science. Then it went artificial intelligence, ai that now you're seeing a lot of buzz around gen AI and a lot of times what happens is Something gets introduced huge buzz and then people start to go, okay, how can I really use this Right? What can I leverage this for? Like, in this gen AI space? I think a lot of companies now are starting to think about okay, what do I use this for? Right, and it's not.

Speaker 3: 14:50

It's not something you use for every solution, but there's clear use cases of it. You know, I was just looking at a demo the other day of like basically talking to a computer where it makes videos for you or it makes images for you right, that's fantastic. If you're in the ad tech space, you can create ads without any resources. That's huge right. So if you're an ad tech company, you should be investing in gen AI, you should be Doing that research right. But not, you know, not if you're any companies I think like no matter what that technology is within data science, what I've seen is it applies to different companies in different sectors in in very different ways. Wow, all right, why.

Speaker 2: 15:32

That's just a great insights, hisha, I'm let me let me kind of wrap up here and see if I can take, you know, some some three takeaways from for our listeners. You know, I think the first one that I heard which is so great is remember this isn't a magic bullet. It doesn't solve all problems. And have a realistic perspective on on what this is. Second, you know, start with small, start with those little initiatives so you can start to grow and understand where it works for you. Should we be doing this for our company? Are we aligned with this? That is the technology. Really, where do we want to see it applied? And then, third, it's really about the culture, and you've got to have a data focused culture and your firm has got to understand and embrace that there is a culture shift if you're really going to take advantage of technology in this space. You equally have to shift the way that you think about data. How'd I do there? Is that a good summary?

Speaker 1: 16:25

No, I thought that was good. I think that I learned a lot. That there is the collaboration is really important to your last point of being there. Wow, I think we've brought it up a level having a doctor here. Thank you so much, ishim, for being here today and enlightening all of the quicksters out there, so we appreciate you, thanks.

Speaker 3: 16:48

Absolutely. This was really fun.

Speaker 2: 16:53

Hey Susie, I was wondering if we were to have a mascot for our podcast. What animal, fictional or real, character or not? Fictional means not real. Which would you choose? And Regis Philbin is chosen.

Speaker 1: 17:13

Okay, so my mind went all through the zoo just because it just allowed me. I have to like kind of do that, and so what came to mind is a mountain lion, a mountain lion, and fantastic and and you that obviously I have to explain why well, there's a bit of like they're. They're a beautiful animal and so they're interesting to kind of observe and they're unexpected, like they're what they do is unexpected and they're also kind of feared Because you don't know where they are.

Speaker 1: 17:53

And so I like the the a bit of Intrigue and surprise about them and their strength, like I just think they're so cool, like I love to watch them walk and just see that, that I do that very often, but I always feel like when I'm on trail some one of them was watching and just making a choice like easy, easy, no way.

Speaker 1: 18:16

We're just gonna leave her alone. We'll wait till there's somebody that's challenging, that goes by, but I I just think there's something that's really captivating About them and interesting and that would make a good mascot for us, because they are clearly captivating, intelligent and Powerful yeah, and I think it would draw attention.

Speaker 1: 18:37

I don't think that's a. I think that that it's a good one. It just reminded me of us. We hope that we're being intriguing and people want to be a part of it, what we're giving them a strength. Thanks for listening to this week's episode of quick take, where we talk about the questions that are on the mind of executives everywhere. Connect with us and share what's on your mind.

Speaker 2: 19:01

You can find us on LinkedIn, youtube or whatever nerdy place on the internet. You find your podcasts. Our links to the show are in the show notes.

Speaker 1: 19:09

We appreciate you.