Katie Sanders: 00:00:00

You had mentioned something like you've had data for 40 years. I think a lot of people have had a lot of data that they've done nothing with or they don't have the right data or they don't have any data.

So we kind of can help with any of that.

Zac Darnell: 00:00:21

Welcome to behind the Product, a podcast by Sep where we believe it takes more than a great idea to make a great product. We've been around for over 30 years building software that matters more.

And we've set out to explore the people, practices and philosophies to try and capture what's behind great software products. So join us on this journey of conversation with the folks that bring ideas to life. Hey everybody. Welcome back to behind the Product.

I'm your host Zach Darnell. Joining me today is my co host is Raman Ori, our president CEO. Now welcome back to our series focused on data.

On this episode we're talking with Mark Daniel Ward, the executive director at the Purdue Data Mine, along with their COO Katie Sanders.

Now this program puts student teams from various disciplines to work on real data problems for real companies, not toy projects, not simulations, actual production grade work for partners like Rolls Royce, Cummins, Corteva, Becks, many others.

And over the last eight years they've put over 7,000 students through the program, expanded to 70 plus universities and are now running 92 active corporate projects this semester. And of course many of which involve some sort of AI. We covered a lot of ground in this show, so let's get into it. Hope you enjoy.

Well, maybe tell us a little bit about. I mean I read your LinkedIn profiles a little bit on the web. Tell us a little bit about you guys and the data mine.

I could read the blurb but I think it'd be more interesting just level set. For anybody listening to this that's never heard of you guys, tell us a little bit about you and Data Mine.

Mark Daniel Ward: 00:02:01

So I'm Mark Daniel Ward. I'm the executive director at the Datamine.

This is the eighth year for our organization and before our organization formed there was a predecessor where we had learning community with sophomores.

And now we're serving graduate and undergraduate students in the whole university and something like 70 other universities that aren't Purdue, you know, and then just this wide array of partners in industry. It's been a ton of fun to get to serve a lot of people.

Zac Darnell: 00:02:27

I love that. So Katie coo, I can make a lot of assumptions about those three letters.

Katie Sanders: 00:02:36

I joke that I basically get to be in everybody's Business all the time. I dip my hands into a little bit of everything. Operations, data, science, and the partnerships. So yeah, I'm in everybody's business.

Zac Darnell: 00:02:50

It sounds a lot like our CEO Tracy. She's awesome and has her hands on a lot of things around scp, thankfully.

So, so I guess the partners, I mean that's really kind of like other than students, obviously the partners are the other half of the program. Tell us a little bit about what it means to be a partner. I saw names like Becks and Cummins and others on the list with some of the past projects.

So what does that look like for you guys?

Katie Sanders: 00:03:12

Our partners, you know, sometimes we seek them out, sometimes they come to us. A lot of students and partners both alike. It's word of mouth, people talking about their own experience they've had with the data mine.

To be a partner, you just have to have a data science problem. We try to work through what that looks like with the partner, walk with them, Project charter, make sure we get students on the project.

We're willing to work with whoever has a problem in the data science realm.

Zac Darnell: 00:03:42

So there's not like, okay, they must be a Fortune 500. None of that matters as much. It really just the data science is.

Mark Daniel Ward: 00:03:50

The anchor point and data driven work for sure.

Zac Darnell: 00:03:56

So this is a very interesting moment for scp. So we're emerging a new data practice as a business since December of 25 here.

I would say that we've been doing data for almost 40 years, as long as we've been in business more operationally focused around software, and now it's more data as a service, data as a product. How do you guys define the word data? Service data capability? Because everybody I ask has a little bit of a different answer to that.

Mark Daniel Ward: 00:04:25

Well, at least the differentiator for us on the academic side of things is, as you kind of alluded to as we were walking in, we're working on real world problems. And it's not toy problems or practice problems or getting students ramped up for real world work.

It's that the students are actually working on the problems that the companies already have. That's kind of the hallmark of our business. That necessarily means what the students are learning is just changing all the time.

And the types of problems that the companies are facing is changing all the time. But even beyond the data, I think another hallmark of data mine is the students are learning about the domain that they're working in.

They'll be working alongside people in digital agriculture and aerospace, in pharma and computational drug Discovery, they'll go out on the manufacturing floor. They're learning from people who are 20 and 30 and 40 years ahead of them in their career. But then they have this really strong data science talent.

And now they just need to learn the domain and think about who's going to pay the bills for them as their career progresses and all of the culture around that.

Raman Ohri: 00:05:29

Now you've made an intentional decision. Like, the student participants in the Data Mine are not strictly data science applied math. They're every discipline. Right.

Can you talk more about that? What drove that strategy?

Mark Daniel Ward: 00:05:44

Maybe that was serendipitous. I mean, we had this statistics living learning community before we had Data Mine.

But the genesis of that idea was even the students came from all these different majors on campus and they were working with faculty from lots of areas on campus.

And then as we morphed into this larger program called Data Mine, people from all these disciplines wanted our students and students all over campus seemed to want to join. Katie said when we first came in that a lot of it is word of mouth. And that's true with the students too.

You know, they just have an awesome experience and then they, they tell their peers. People come to Purdue already having heard about Data Mine in high school and then knowing they want to get involved.

Katie Sanders: 00:06:23

I think another piece of that too is people look at data differently. A marketing person might not be a data scientist, but they have experience in marketing that they can look at a problem one way.

Engineers look at data differently than data scientists. So you can get different input from different groups of people, which I think makes the project stronger.

Raman Ohri: 00:06:48

Do you specifically target data science projects or do you end up finding opportunities that kind of span data engineering, data science and other.

Mark Daniel Ward: 00:06:58

We say data driven. If they're data driven problems, that's what we're really looking for.

Katie Sanders: 00:07:04

I would say they span. Like I will use Microsoft as an example. You can see the Minecraft work that they've done on our website.

It's more marketing data driven than it's actually data science.

Zac Darnell: 00:07:16

Y well, and I mean like, that's even like an interesting conversation going back to eight different ways to define a data problem. You know, air quoting constantly when I'm talking about this stuff. Like data science is one of those for us as a business.

We would say you probably, you may not want to hire us for data science because, you know, we might assume that that requires deep domain expertise to be really effective at. And yet at the same time, there's a lot of readily available, tried and true algorithms that you can Pull off the shelf for a lot of common problems.

Like that's also data science work. You don't have to necessarily have deep domain expertise.

Like do you find that your folks tend to fall into one of those two buckets or do they stay more in that? Okay, there's 10 algorithms that I can pull off the shelf to solve this kind of a problem.

We're going to work our way through it to find the best fit. Or are they more on the. I'm going to go custom made tailoring this algorithm to this problem. Like where do you guys tend to find your projects?

Mark Daniel Ward: 00:08:13

There is some R and D where students are trying some algorithms and seeing what's best practice or what's fastest or most efficient or whatever. But I do think a lot of our companies want to keep everything proprietary and we frankly don't publish very often.

We're usually doing work that can't be shared with the public and as a result it's just a team needs a solution.

So it could be something off the shelf as you alluded to, but it's usually something really customized and maybe there's a partner for the partner or a third party there behind the scenes. We may never even meet them sometimes. But yeah. So lots and lots of customized solutions.

Raman Ohri: 00:08:51

I hadn't considered the not being able to share the work. So produce obviously I can't remember the nomenclature like a tier one research school.

Do you find sometimes your students are kind of torn like yes, I would like to do the data mine, but I won't be able to talk about what I do or I won't be able to publish. Or have you found a way to.

Mark Daniel Ward: 00:09:08

Navigate that certainly positively impacts when they do go publish about other problems that they work on and impacts where they get hired. And I think we're transparent about it.

Coming into data mine, we're not endeavoring to publish in top tier data science journals or AI conferences and things like this where a lot of the really cutting edge work in pure data science principles is happening.

It's more like we're just out there consulting and getting stuff done with this huge class of early career people who are just really eager to go get good jobs and do the work.

Katie Sanders: 00:09:42

I mean you can see the projects online that we do. All of those, they undergo a legal.

Mark Daniel Ward: 00:09:49

Like a review or a vetting, you know, for several weeks before they go.

Katie Sanders: 00:09:53

Out, before we have our symposium.

Mark Daniel Ward: 00:09:55

So think about our organization. Well, for sure, make sure we're not.

Katie Sanders: 00:10:00

Saying anything we're not supposed to.

Mark Daniel Ward: 00:10:02

So if you think about our Organization though We've had roughly 7,000 students throughout the year. Okay. But they're also early in their career. We only started eight years ago. Right.

So there are managers out there and there are people that already are early career AVPs and things like this, but most of our alumni are just out there actually getting the work done.

And so when we're visiting companies, we almost always have some alumni who are there and we're like, just tell us about how your work's going, you know, what's your team like?

You know, how's the mission of your org changing as the data sciences are impacting your work, maybe there's some ways that we can help and you can lead a team.

And because they're not a manager yet, lots of our students are getting these really early career leadership experiences that have landed out in industry and that's where we're really starting to see this snowball effect.

Zac Darnell: 00:10:50

So the program's been around for eight years.

Like you say, think about the last three years in this explosion of gen AI and LLMs and that impacting just about every aspect of everybody's world for the most part. How have you seen that shift for your guys program, the impact of what that's looked like both from a student and even from a project perspective?

Asking for a friend.

Katie Sanders: 00:11:14

It's funny, we actually did a breakdown of how many projects include AI. We have 53 of the 92 include AI now. So we're having to up our skills so that we stay relevant. Right.

Zac Darnell: 00:11:35

And are you seeing them in kind of the gen AI LLM space or.

This is like we've been doing traditional AI, NLP and planning and scheduling optimization for longer than the last three year hype cycle has been around. But this is all that people really want to talk about, seemingly the new hype cycle.

Have you, have you seen a disparity between those two things or are you seeing kind of a similar trend?

Mark Daniel Ward: 00:11:59

Yeah, we've definitely done NLP projects for quite some time and there were a lot of LLM projects this year.

I think people are already getting excited next year about how all the LLMs are going to have impact with streaming video, streaming images, you know, all the generative AI, not just for tech space, but for videos and images and so on. Right. We already kind of see that's going to be a dominant theme next year.

nows what they'll be doing in: 2028

And you kind of alluded to that with SCP as well, right?

Zac Darnell: 00:12:42

Yeah, a little bit. Yeah. It's been interesting even this year.

Kind of the new round of folks coming out of school and kind of getting on their first team, their first project, and seeing the difference in the way even they approach using some of these AI tools versus some of the folks that have been here for the last few years learning and adopting them. But 10 or 15 years into their career, they have that foundation like you talked about.

It's just, it's different to kind of see the, the two different lenses kind of using the same tools. There's some nuance there. There is, there is some impact. You know, it's not in a bad way. It's.

It's great to have that collision of thought and perspective and experiences. But it definitely is, as a recent realization, like, oh, you think about this differently. That's interesting.

We might need to think about the way that we do some of that foundational work. You know, it's interesting, you know, when.

Mark Daniel Ward: 00:13:32

A student's on one of our teams too. I mean, Katie kind of alluded to it earlier. There's students from all these different backgrounds and they're different level.

There's sophomore undergrads up through graduate students. So when you've got 12 students on a team, no one student is just doing all the work. They have their little piece of it that they kind of own.

They feel confident about it and they've contributed it and you know, it's like their stake in this larger effort and so they feel really valued as a result. Students don't always feel that way in college or even during their graduate school experience.

And I think in data mine, there's this sense of, of culture and support and length of the program gives them even some freedom to fail. It's a very comfortable while still being challenged environment.

Raman Ohri: 00:14:16

We've talked a lot about the students. Let's think about the companies for a minute. What does that look like?

Say somebody's listening to this podcast and they say, well, I have a lot of data science problems or data opportunities. What does it look like from kind of start through delivery? What is that experience for them?

Mark Daniel Ward: 00:14:34

You want to take this?

Katie Sanders: 00:14:35

Go ahead, give us a call.

Mark Daniel Ward: 00:14:39

I was thinking the same. Our email is State of mind at Purdue Edu. Hello.

Zac Darnell: 00:14:44

Let's go.

Katie Sanders: 00:14:46

But no we sit down with the person, we can talk through what those problems are, what might be a good fit. After we decide what that might be collectively because we pull in a data scientist to help us with that as well.

We have four data scientists on staff that can help us. Then we come up with a project description and we kind of figure out what kind of students you're interested in working with.

Then we get the students and we work on a project charter and we get going.

Zac Darnell: 00:15:18

Now, I saw that there's a list of project opportunities that I see on your website. Now. Is this okay? Well, students get to rank order the ones that they find that they want to work on.

And the company is kind of waiting like, well, I hope my project gets picked up by the students. And like, is there a. If I submit a thing, if my thing doesn't get chosen, then, well, unfortunately I don't get to work with the data mine.

Like, how does that part work?

Katie Sanders: 00:15:42

No, we'll find something that works.

Zac Darnell: 00:15:44

Okay.

Katie Sanders: 00:15:45

Yeah.

Zac Darnell: 00:15:45

All right.

Mark Daniel Ward: 00:15:46

Every team was full this year.

Zac Darnell: 00:15:48

Yeah, they're all full.

Katie Sanders: 00:15:49

Yeah. Also, you know, we have a vast array of students who want to be a part of our program.

So only about 50% of our students are in our corporate partners program.

The rest of them are in our seminar which teaches them introductory skills to data science so that they can use them when they're able to get on the corporate partner project.

Zac Darnell: 00:16:09

So I mean, you effectively have a two sided marketplace play. Right. Uber for data science.

Katie Sanders: 00:16:15

I like that. We'll go with that.

Zac Darnell: 00:16:18

It sounds like you've got a plethora of student availability and maybe looking to grow the opportunities that they get to work on. If I'm hearing and understanding that.

Mark Daniel Ward: 00:16:27

Right.

Zac Darnell: 00:16:28

And the job is to match those opportunities and make those work well.

Mark Daniel Ward: 00:16:32

Oh yeah.

Katie Sanders: 00:16:33

So the students get to self select.

Zac Darnell: 00:16:35

Okay.

Katie Sanders: 00:16:35

Most of the time they get to self select on which project interests them the most. To answer that question. Sorry, that was like the long winded answer.

Raman Ohri: 00:16:45

So if I'm considering being a partner, like, how do I get the most out of this? How do I to the suffer success?

Mark Daniel Ward: 00:16:55

I could frame that. It's like a relationship.

And we know that if the partner has a good experience the first term, they're going to come back year on year on year on year.

So we try to pick a project that's really interesting, both for the students who are going to be involved, but will be a lot of ROI for the company too. And something that we know will be kind of foundational.

The students will be able to build on it likely in years ahead or it will spawn other ideas as it goes along. We try to paint in their mind, hey, we're going to go and sign a five year contract.

It doesn't mean you have to come back and work with us in the second or third year and so on. But we've already done all the legal framework and we're building the grounds for the relationship in that way.

Students also tend to come back to a team and so we do, I won't say micromanage, but we really do try to find the right students for those teams the first year and then we know that there will be many returners and also people will hear about the experience and will be naturally drawn to it after the first year we work together.

Raman Ohri: 00:17:52

So the first time a partner shows up, do they tend to have like, nope, this is the thing that I need solved or do they have like a shopping list and you matchmake with them or neither.

Katie Sanders: 00:18:04

Depends.

Mark Daniel Ward: 00:18:05

I mean often they'll have somewhere somebody's keeping track of all of this backlog of work that's just not getting done or it's going to be done offshore or it really has high roi, but there's just no one there to do. It could be because of people or it could be tech skills or whatever.

But we usually do encourage them pick something that's really high return on investment, but no one's going to die if it doesn't get done. And we know it's students working on it, right? Not professionals.

We like problems where maybe there's a little bit of tunnel vision in the past and people want a new view of things. Students will try something that was unexpected.

What will tend to happen too is they'll have a long list and we'll pick one, but we'll know that going in there will be two the next year and four or five the year after that. And again it'll kind of just continue to grow as naturally.

Katie Sanders: 00:18:51

So you had mentioned like what makes a good project basically for a partner.

And so the way that the corporate partners program works is there's a two hour lab which is led by a TA and there's a 50 minute mentor meeting each week. Having that mentor be engaged and interested in the project makes the project really good for the company and for the students.

Having a great TA helps as well. A lot of our TAs have been data mine students so they kind of know what to expect, they have a background.

So our hope is that that relationship and the communication makes for a really great project.

Zac Darnell: 00:19:28

Now does that ta kind of sounds like they're kind of like functioning like almost like a project manager. Is that kind of their role?

Mark Daniel Ward: 00:19:33

They're the Scrum master.

Zac Darnell: 00:19:35

Yeah, that makes sense. And then the, the mentor that's coming from the, the. The partner company, they're the product owner. Perfect. Okay.

Mark Daniel Ward: 00:19:43

Yeah.

Zac Darnell: 00:19:47

Okay, that makes total sense. So the, the. There is some level of, hey, guys, for this to go really well, you need to have some time allotment and some engagement.

You know, minimum 50 minutes a week, it sounds like, to kind of be that role.

Katie Sanders: 00:20:01

And the project charter helps too. Right. So then the mentor can say, this isn't working. This isn't what I asked for. Can we go back to the drawing board?

Mark Daniel Ward: 00:20:09

We train the students with Agile as well, and we tell them going in, I mean, this is kind of where we're aiming to be in nine months, but we know as we go along, we're going to pivot and turn and do unexpected things.

Zac Darnell: 00:20:19

Yes, yes. Having. Being in a role where, hey, what's going to happen in nine months? I don't know.

We hope to have this kind of outcome, but yeah, there's a hundred ways we can get there. Speaking of outcomes, like what the end of the nine months they get done with this project, are you feeding the pipeline for these companies?

And hopefully these students get job offers and like, that's the ultimate goal here.

Mark Daniel Ward: 00:20:44

It's one of the goals. I mean, nobody has to promise to hire anybody.

Zac Darnell: 00:20:48

Sure.

Mark Daniel Ward: 00:20:48

You know, but a KPI might be okay. Well, this product's going to perform better than something we've built on our team in the last year.

Or we're going to compare to something that we built recently. And this one's. It's innovative in some way. We did something different.

Or maybe we took on some technology we've never tried before, or maybe we didn't know which technology to use and the students did all of the R and D and came back and gave us feedback at a price that would be way cheaper than a consulting agency or something like that. So, yeah, the hiring is important, but I think also again, that relationship with the company is really paramount importance.

Zac Darnell: 00:21:26

What favorite project so far in eight years?

Katie Sanders: 00:21:30

I like all the chatbot stuff that the students are doing.

It seems like a lot of companies are interested in having their own internal chat so they can just ding, ding, ding, and then query around it right there. I love that. And I love that the students can figure that out. I mean, when I was a student, I wasn't doing that.

Zac Darnell: 00:21:48

Not me either.

Mark Daniel Ward: 00:21:51

We do a lot of small and mid sized businesses. I mean, of course there's plenty of Fortune 500 companies there for sure, and there's some startups, but in between there's this murky middleware.

You've got a small or mid sized business and they're just so eager to have really talented early career people. Data Mine's a great way to do that.

So I find the most rewarding projects are the ones where the manager comes back and is like, well, I never would have found this student without Datamine or I never would have gotten this thing done on time or I wouldn't have thought about that approach without these students. It's really satisfying. Some of the outcomes we have, I.

Raman Ohri: 00:22:27

Feel like it's not based on things you said. It's not explicitly the mission, but this idea of placement and real world experience for the students.

I'll speak only to the computer science area, but it's been brutal, brutal for graduates and even just to get internships. And I've talked about this many times, like just little old Sep.

In the last three years, we've seen our intern candidate flow go up 15x 15x like it's insanity. Like really good students from really good schools struggling to find.

And so, I mean, part of what I found, like what enamored me with the data, mine was you're bridging it in a different way. Okay, maybe a company's not willing to commit to the internship, but it is a different way to establish a relationship.

Both help the student and also audition them. And are your students finding that this is helpful?

Katie Sanders: 00:23:22

Yeah, I think one of the greatest things is you have nine months to meet eight to 12 students to have them learn about your company. They're already gonna be embedded in it in some way, so they have some knowledge.

So if you choose to hire them or you choose to have them as an intern, they're already gonna know something. It's gonna be less training for them, for the company. So I like that.

Mark Daniel Ward: 00:23:46

I think one weakness we have on our side being academics is even in the best computer science programs in the country. I mean, Purdue has an amazing computer science program, just amazing program.

And the skills the students learn there are, I mean, just some of the best in the whole country. But I often challenge the students. You know, who's going to pay your bills in five years? You know, how are you going to get that first mortgage?

Like, what kind of team are you going to be embedded in? And when you really ask even the most technical student, what do they want to spend their career doing.

They'll know the tech stacks they want to work on. They'll know what kinds of applications as an example or what they enjoy coding.

But they won't have always thought ahead to the people and the nature of the work and the mission of the company or what's their dream job. And if they have thought about their dream job, it's usually one of the huge tech firms they haven't thought about.

I really want to spend my life in Carmel, Indiana working at an SEP or I want to devote my life to computational drug discovery. I want to be out on the manufacturing floor at Allison Transmission.

You know, they're not always thinking with the sense of place and like local economies and even the domain they're going to work in, you know, they're just still often, I mean they're just, they have that early career mentality. Right. So they don't know what they don't know yet.

Katie Sanders: 00:25:03

I think about one student, he was really interested in working for one of the big companies and then he did a project with a mid sized company that he learned about through the data mine and he said, I didn't even know that I could work in a company like this and do this kind of work. So I think it brings some visibility to smaller companies. So I love that story.

Zac Darnell: 00:25:23

No, that's, I mean that's a really good point. There's different constraints at different sizes and that comes along with a very different work environment.

Mark Daniel Ward: 00:25:30

A lot of ways we should mention our friend. I think we can tell Mike Douglas from Raytheon or now rtx. Mike's been with us almost since the beginning.

He comes up to campus and he has lunch with the students and goes out on bike rides with them and is really immersed in their experience. And if Mike's listening, he jokingly says, oh, I'm going to quit if my boss ever takes data mine away from me. I mean he loves it.

But one thing he's emphasized to us is it's one of the most rewarding parts of his job. And I think we see that with a lot of our mentors. You know, they still got their other stuff to do the other 39 hours of the week.

But for this one hour they get to be with these really early career people who, okay, they're not reporting to them technically. Right. The company's not going to live or die based on their experience. But it's, it's a great way of giving back.

Zac Darnell: 00:26:16

Yeah, I mean that there's so many different aspects of this. You Know, again, if I go, I go back to the Uber for data science connection. Just think about it as a business model.

You know, like there's, there's a lot of challenges to starting a two sided marketplace business. You've got the consumption side and you've got the production side.

As you guys think about like the challenges that you've had or maybe are currently having, like what does that look like today?

Is it, is it simply looking for more consumers or is there something else at play that's making, I don't know, that you're losing sleepover or that you're thinking, oh man, how are we going to get that done? I don't know. I find that interesting, the kind of the hurdles that a lot of companies and organizations have to overcome.

Mark Daniel Ward: 00:26:59

illion gift five years ago in: 2020

And no doubt about it, I mean, even more than the grants we've had from the federal government and from our very generous partners and so on, and our Lilly Endowment grant ends next year. So the university just candidly is thinking, well, what do we do with the data mine? This is a model unlike any we've had before.

But earlier we were mentioning about Purdue's role in the state. Purdue is also the land grant for the entire state of Indiana.

So all of these small and mid sized colleges are also looking at data mine mod and how it fits into their local economies and departments and so on. And so I think, at least for me, the thing I'm most excited about is, well, what's the future going to bring?

. What does that look like in: 2030

Zac Darnell: 00:27:58

Yeah. Any guesses? Any crystal ball reading?

Mark Daniel Ward: 00:28:02

Golly. One fun exercise we often do is we're think about, okay, well how quickly are we going to grow, what resources do we need to get there and so on.

And it's just an uncertain time, right? I mean, so we're pretty resilient in our team and we just kind of take things as they come and we do the best we can to serve our partners.

And one person early in the life of data mine said, if your students have an amazing experience and your partners have an amazing experience, you're not going to have to worry too much about your funding. It's going to come in at the rate it comes in and your organization will grow.

Katie Sanders: 00:28:37

If you know anyone with a crystal.

Zac Darnell: 00:28:39

Ball, that's why I'm asking you. I don't have one myself. I pretend to think I know what might happen in the next month, let alone the next five years.

But it's fun to sometimes dream a little bit and kind of see where you're at and cast a vision for the future. So I always think that's an interesting conversation. I love it.

Raman Ohri: 00:28:58

Well, I have one.

Zac Darnell: 00:28:59

Yes.

Raman Ohri: 00:29:01

I don't even know that it's a prediction so much as maybe an assessment.

Zac Darnell: 00:29:04

Of now, but hot take, hopefully.

Raman Ohri: 00:29:06

I don't know how it is, and I don't know how original it is. I'm sure others have thought it too. But increasingly, I think there are very few businesses not permeated by AI. And if they're not yet, they will be.

And what I mean by AI could vary wildly. Are you talking about Waymo? Are you talking about a pure SaaS business? Right. Like, it's all over the place.

But increasingly, I think the only differentiation you will have is what data do you have that nobody else does? And so if you're an organization today that's thinking about your future and strategically, like, what are you doing to lay the groundwork for that?

Where do you have some sort of untapped resources? And I think programs like this could really help companies start to unlock that. So I don't know.

I guess I just did a commercial for you, but my prediction being the companies that understand. So I'll get to the prediction.

The companies understand that the best will be the best position three to five to seven years from now because of the groundwork they do now around their data.

Zac Darnell: 00:29:59

Yeah, you kind of talk about the kind of spectrum that we've seen in our customers over the last few years in the realm to AI. Hey, I want to do some AI. It's like, okay, great. Come to find out you guys haven't been collecting any data.

It's really hard to machine something that doesn't exist, or you haven't been collecting the right data, or it's not ready to be machined. There's a whole plethora of data problems before you can do the cool stuff. What do you guys often see within your partners?

Actually, I'll turn that commercial into a question.

Katie Sanders: 00:30:33

We see all of that. I know you had mentioned something like you've had data for 40 years, right?

Zac Darnell: 00:30:39

Yeah.

Katie Sanders: 00:30:39

I think a lot of people have had a lot of data that they've done nothing with, or they don't have the right data or they don't have any data. So we kind of can Help with.

Mark Daniel Ward: 00:30:51

Any of that, all the workflows and just the project oriented approach.

I read this thing in the Harvard Business Review recently where it was talking about how in many economies, in many sectors of industry, things are becoming less and less driven forward at an enterprise level. The CEO or the COO or the CTO comes and says this is how we're all going to be aligned and things are much, much more project oriented nowadays.

And kind of our whole operating model really fits in well with that. If there's projects that are excellent and rise to the top, then okay, great, then what? What do you do next? What, how do we build on that? Yes.

Zac Darnell: 00:31:28

So, all right, so it's like a third, a third and a third, all the problems. I love it. Very evenly split, evenly distributed. Okay. So, you know, we could, we could dream about the future.

We could, we could think about high level outcomes when we get super myopic and tactile for a hot minute. You know, again, we're talking about our.

I don't want to make this about us, but for context, starting this data practice officially as a business, you know, one of the things that we've been thinking about are the various roles that might be needed that are new to us. Are folks in the building? Are we going to need a new prefix? For example, instead of being a software engineer, you can be a data engineer.

Well, okay, if you're a data engineer, what does that even mean?

What have you guys seen across customers, regardless of the problem area, for the roles that they need to accompany some of this data science work that they're, that they're needing to embark on? Yeah.

Mark Daniel Ward: 00:32:20

I mean, one thing we see, I mean like when we talk about the Fortune 500 companies, I just, it never ceases to amaze me that every single company has both AWS and Azure investments. And it's not just like a few people in the company using aws, if you're using Azure, I mean, it's, it's like Coke and Pepsi.

I mean, honest to goodness, they throw,.

Zac Darnell: 00:32:37

You know, GCP in there and you got Dr. Pepper. Right.

Mark Daniel Ward: 00:32:39

For sure. You got their GCP in there a little bit. Right. So it goes all the way from the cloud level down to the hardware level and APIs.

Of course, you've spoken to AI Innovations and LLMs and so on. Right? I mean, so I would guess for an SCP with. You said there's 100 odd people here, right?

Zac Darnell: 00:32:59

Yep. About 180.

Mark Daniel Ward: 00:33:00

Yeah. I mean, it maybe, I mean, I don't know anything. Right. But it just doesn't hurt to anticipate.

Okay, well, we're gonna need to get three more Azure developers in here and five more AWS developers.

But you know, we had some exits in our hardware team or, you know, maybe it's time to do some training with our API team or something like that, you know, and just, just be kind of thinking ahead to, you know, what kinds of projects are you, are you surfacing, you know, what contracts are.

Zac Darnell: 00:33:27

You winning and yeah, it's been interesting, you know. Yes, some, you know, our customers seem to be split into one of those two.

Every once in a while we get, you know, multi cloud and I want to be in both. You know, then there's the, you know, we want to do and use Snowflake or we want to be pure Azure. Whatever, whatever it is.

It's been interesting to see kind of these specialty roles versus the more generalist roles and kind of how those have just the mindsets inside of those, those different kind of focuses. There's been no consistency that I've personally seen so far.

So it sounds like within the customers and the programs that you guys are working in, it's all of the map.

Mark Daniel Ward: 00:34:06

Just because we take on such broad book of business as well, I think, you know, we're also a little bit agnostic in that we have this. I mean, we have 2,000 some employees in a sense. Right. You know, so.

Zac Darnell: 00:34:17

No, that makes sense.

Mark Daniel Ward: 00:34:19

We're just a large organization.

Zac Darnell: 00:34:21

Okay, well, back to spicy takes. Any, any hot takes.

Like, is there, is there a belief that you have inside of kind of the focus for the data mine that you think is antithetical to a lot of the way that your customers show up.

Mark Daniel Ward: 00:34:37

I could say again, one other hallmark aid of mine we haven't commented on too much is like, I love hanging out with our colleagues in the School of Agriculture at Purdue as an example, and technology or in liberal arts as an example.

Like some of the most satisfying partnerships and student experiences and outcomes and so on are tied to the fact that we are kind of like we alluded to earlier, we're just, you know, we're very domain centric and we like to think about the impact we're going to have, aside from the tech point of it, in, in a domain, you know, out in the field on the manufacturing floor and so on. So yeah, it's, it's easy to say that and okay, yeah, sure.

But I think that's, that's what kind of sets us apart is we're really genuinely very interdisciplinary in the kinds of problems we can tackle.

Zac Darnell: 00:35:23

That's fair.

Mark Daniel Ward: 00:35:24

Yeah, yeah.

Zac Darnell: 00:35:26

Okay.

So, you know, we were just talking a little bit ago before we turn the mics on, the school year is pretty much, you know, quote unquote over at this point.

Mark Daniel Ward: 00:35:33

It's next August already.

Zac Darnell: 00:35:36

So, you know, obviously we're recording this In December of 25th, if we forecast to August of next year. What's going on? Yeah, I think you did, I read, right, that you guys are building and expanding more space to accommodate students and facilities.

Some of that's going on right now. Like what else is looking or is coming up very soon in August and the school year for 26, 27.

Mark Daniel Ward: 00:36:01

I mean, Purdue's been very good to us in that we are building a new building with roughly 900 more students, I think, and the board of trustees has designated it for the Data Mine. And we were involved in the process of designing the building.

So rather than having our offices there because we don't know how big our team will be in five years, we've put meeting rooms in the first floor. So it's very student centric. You know, everything takes place in the student's, you know, community.

It's all of Data Mine is in the learning communities in Purdue. You know, that's pretty central to our organization. We're right across the street in the co working space where the companies reside on campus.

Katie Sanders: 00:36:38

So we also have our symposium in April. The students showcase all of their projects. It's really fun. Takes place in the CO REC.

It's on April 29, 4pm Anybody can come watch the students present, learn about what they've done. It's a great opportunity for people who are interested in learning more about the Data Mine.

Mark Daniel Ward: 00:36:58

We can put the link for them from the previous years if people want to see what we've done in the past.

Zac Darnell: 00:37:03

So anybody that's open to the public.

Raman Ohri: 00:37:05

Yeah.

Mark Daniel Ward: 00:37:05

Oh, it's like a party.

Zac Darnell: 00:37:07

That's great.

Katie Sanders: 00:37:07

It's fun.

Mark Daniel Ward: 00:37:08

It is fun.

Zac Darnell: 00:37:09

Okay.

Mark Daniel Ward: 00:37:10

Yeah, yeah.

Zac Darnell: 00:37:11

How about how many projects are going on at any given time?

Mark Daniel Ward: 00:37:15

Well, all our projects last the whole academic year, right?

Zac Darnell: 00:37:17

Sure.

Mark Daniel Ward: 00:37:17

I think Katie alluded to there's about 92 credit based projects.

Zac Darnell: 00:37:22

Okay.

Mark Daniel Ward: 00:37:22

And then there's our sponsor research we don't talk about publicly, but sure, but people from all of these companies will come and meet the students and take them out to dinner and kind of celebrate the work from the whole year.

Zac Darnell: 00:37:33

Okay, sorry I missed that. 90 Is what's going on concurrently.

Katie Sanders: 00:37:38

So we actually are starting 10 new projects in the spring too, which we don't love doing semester long projects because it's hard to see results in that short amount of time. But it's a good Runway for the next academic year.

So students could get started in the spring and then the company can come back and work with us in the fall and next spring. So we're starting 10 new ones this spring.

Mark Daniel Ward: 00:38:02

Okay, so an example would be like in October, we met with cnh. They were on campus and they said, well, why don't you come up to Racine and just meet with us and we'll, you know, like brainstorm a little.

Well, if we, if we start in October, everything's already launched, but there will certainly be students who didn't have a project in the fall and just can't wait to get started next year. So CNH will offer a couple projects in spring and it also helps them get ramped up and going before, since we haven't worked with them before.

Zac Darnell: 00:38:29

That makes a lot of sense. Yeah. Well, that's exciting. Coming up for the spring, is there anything else that you feel like either a partner or a student?

Probably our audience is more on the partner side of things. So what would you want them to know about the data mine that maybe we haven't chatted about?

Katie Sanders: 00:38:45

I think there's a lot of different reasons companies choose to come to the data mine. Some it's for recruiting purposes.

I would say probably the majority they want students, but some come just because they have projects that have been on the back burner that aren't mission critical, but they really want to get it done. I think there's a lot of things that we can do to help. We're not super expensive either.

I think it's a great way to get connected with the university too.

It's a good entry point and we love to meet new people and introduce the students to those people as well, so that the students and the companies both have a good experience.

Zac Darnell: 00:39:20

Sounds like it's great value.

Katie Sanders: 00:39:22

Well, I mean, I think so, right?

Zac Darnell: 00:39:23

Yes, great value.

Mark Daniel Ward: 00:39:25

It'll lead to bigger things with university. I alluded to our sponsor research, which of course is, you know, at a larger scale as faculty and departments and so on.

But some of our engagements have even led to master research agreements with the university where all the legal stuff is articulated up front and then over a period of time, if something comes up with the company and they want Purdue to be a partner, you can just write down a one page scope of work and dollar amount and personnel and execute immediately and those kind of relationships. The university really genuinely values you know, then the company and the university feel like we're really in it together.

Zac Darnell: 00:40:02

That's perfect. All right, Ramen, any last questions here before we wrap up? Appreciate you both.

Mark Daniel Ward: 00:40:08

Thank you.

Zac Darnell: 00:40:09

Mark and Katie, thanks for. Thanks for driving down. Yeah.