The Data Cloud Podcast

Unlocking the Power of AI with Dan Wright, President and COO of DataRobot

Episode Summary

This episode features an interview with Dan Wright, President and COO of DataRobot. Dan has deep experience driving growth and operational excellence at disruptive technology companies like DataRobot, AppDynamics, LogDNA, People.ai, and more. In this interview Dan talks about machine learning, neural networks, how to get value from AI systems, and much more.

Episode Notes

This episode features an interview with Dan Wright, President, and COO of DataRobot. Dan has deep experience driving growth and operational excellence at disruptive technology companies like DataRobot, AppDynamics, LogDNA, People.ai, and more.

In this interview, Dan talks about machine learning, neural networks, how to get value from AI systems, and much more.

--------

How you approach data will define what’s possible for your organization. Data engineers, data scientists, application developers, and a host of other data professionals who depend on the Snowflake Data Cloud continue to thrive thanks to a decade of technology breakthroughs. But that journey is only the beginning.

Attend Snowflake Summit 2023 in Las Vegas June 26-29 to learn how to access, build, and monetize data, tools, models, and applications in ways that were previously unimaginable. Enable seamless alignment and collaboration across these crucial functions in the Data Cloud to transform nearly every aspect of your organization.
Learn more and register at www.snowflake.com/summit

Episode Transcription

Steve Hamm: [00:00:00] it's nice to meet you, Dan.

Dan Wright: [00:00:02] Great to meet you too, Steve.

Steve Hamm: [00:00:03] now. You're out in California and the is in Boston and, that's kind of an unusual setup how do you plug into the company from, from such a distance.

Dan Wright: [00:00:14] Obviously everybody's a remote right now, but we've been a pretty distributed company. Uh, , from the start. So we have a significant office in Tokyo, in Japan, a significant presence in London and, throughout the United States. and obviously the, the headquarters, since the time the company was founded in 2012 has been in Boston.

but we also do have a significant presence in the Bay area where I'm located and over the course of the last year, Acquired three different companies in the, in the Bay area. So we had a critical mass. I also have a personal connection to Boston. I lived there for three years before moving out,  to California, where I'm from originally.

and then, , Jeremy and I, , hit it off immediately and kind of knew that we could work together now withstanding the distance. So it's been really good and I come to Boston when we can travel and he comes here and it works pretty well.

Steve Hamm: [00:01:07] Now you joined data robot as, as president and COO early in 2020, previously, you were the COO with app dynamics, both when it was an independent company and after it was acquired by Cisco.   What's it like to get back into the startup world?

Dan Wright: [00:01:25] it's amazing. , I have a real passion for, company building and also for, bringing new technologies that, , I just have a ton of value to customers, to market. And, , I love that at app dynamics. and , the opportunity to do it again now with a data robot has been really energizing for me, exciting for me.

and , this technology that we have, , is truly unique and, , helps to unlock the power of AI, which , is significant and will become more significant, as we go. So it's been, , great. I jumped, joined the company. January and I've loved every minute since I joined.

Steve Hamm: [00:02:05] Yeah. Is this your first time as an executive of an AI company? Have you had to have a crash course in machine learning?

Dan Wright: [00:02:14] yes and no. So, even when I was at app dynamics, both pre and post mergers, there was a machine learning actually built into our product. So I was, definitely familiar with.  how it works. and then also the power that it can bring to, to customers in terms of, automating, traditionally manual tasks and speeding up.

Uh, both the path to value and to resolving, pain points in the enterprise. And, , so I had familiarity, , at the same time did a robot is a unique company it's, , end to end enterprise AI. It's all we do. And so that is a different ball game. So had a good foundation to build from, but if, obviously learned a lot since joining,  back in January,

Steve Hamm: [00:02:57] now data robots, failure proposition is based on the idea of automated AI. What does it mean to automate AI? If you could walk us through how people use the platform that might help.

Dan Wright: [00:03:10] So it's easiest to describe if you look at the history of AI. So before companies like data robot, All of this was done manually. So people would actually create the machine learning models,  manually, uh,  , coding them. And then, , they would, uh, do, uh, all of the,  tagging and cleaning of the datasets so that they could then apply that too.

the model. And then they would kind of hand off to dev ops and get those models into production. And then every time the data change you had to manually update, the model in each of those steps could take months if not longer to complete. And so data robot, when the company started in 2012 was the pioneer of, first automated machine learning.

So rather than somebody having to manually create the models. Data robot was able to automate that entire, process and take it one step further by I'm taking all of the different potential models, including the best of open source,  data robot, specific models and others, and figuring out what is the best model for that specific dataset.

And it was able to do that within, , minutes versus weeks. And, , with,  a much higher quality of model, you knew that what you were getting was the very best. And then,  what we've been able to do since that time has been to expand, to automate the entire end to end journey that I was describing that was done manually before data robot existed.

So all of the tagging and cleaning of the data. Which again was highly manual before we're now able to do automatically. We acquired a company pick Sada, last year and I've integrated into the platform so that we can do that also when it comes to deploying models in production, we have a, an ML ops offering.

that allows customers to not only automate that process, but then once the models are in production, manage all of them and be alerted if there's any sort of data degradation or drift, so that the models can be quickly updated based on the changes in data and then redeployed so that, , companies can get the value very, very quickly without any disruption.

so we've really automated every step of that. And now what we're doing is actually because we have the end to end platform, we're able to create applications on top, which we've now started shipping with our. Some of our recent releases. And what that allows you to do is as a consumer of AI, get the value in a more streamlined, automated way, which is important because one of the biggest impediments to companies adopting AI in the past has been that it was a highly manual process.

And as a result, it took a long time for companies to see value and then you would lose momentum. So we think by speeding all of that up, it's only gonna accelerate this AI revolution that we're

Steve Hamm: [00:06:06] focusing now on the, the data analyst or the business analyst, what's the experience that they have and the process they go through using your company's technology.

Dan Wright: [00:06:17] So one of the missions that we have and have had from the very start is around democratizing, AI. So the idea is that you shouldn't need to be. A PhD, , in data science to use this technology. And that's important because the supply of, , those types of people, is, , a much smaller than the demand for AI.

And so it's imperative for, just. The state of technology and companies at an individual level to enable the rest of their people. Business analysts are very well positioned to be able to use this technology. Now with the automation that we put in place, we say that all you really need is,  some curiosity.

And then some motivation and you can become successful. And what that looks like for a business analyst is you can literally, , automatically curate the data. we have an AI catalog where you can also select a different datasets. And then you can, literally just press a start button angel, get your, model popped out.

And that continues throughout the entire chain from data to value. So,  it's very quick to, to get to the point where. Uh, you can create a model in production, that's getting value. And then,  for a business analyst, if you're somebody who's accustomed to working with a Tablo dashboard now, we're able to, , basically provide that same visibility.

But use it, with,  predictions predicting what is going to happen in the future versus just reporting what has already happened in the past. And we think for business analysts that has the potential to be a game changer and really allow them to add a lot more value to their companies.

Steve Hamm: [00:08:06] So basically it sounds like one of the things you do is you take a whole process that normally, or in the past it took a data scientist to do, and it might take months now a business analyst can do it and do it in a few minutes. Is that correct?

Dan Wright: [00:08:24] That's exactly right.

Steve Hamm: [00:08:25] That sounds almost miraculous.

Dan Wright: [00:08:28] Well, we actually say it's a, it should. And it does to our customers feel like magic. but the nice part about it is, , because we sell to many of the fortune 500 in the global 2000, it does feel like magic, but the nice thing is if they want get a blueprint of everything that's going on, sort of under the hood and understand.

, in a fully auditable, way, , what, what is actually happening. And then, , that's very powerful.

Steve Hamm: [00:08:52] so they can go back and audit it and look through the process and see, well, this is this, there was this whole automated thing that I didn't see into. I just got the results, but now I can check back and make sure that I understand the process and there's nothing  that like the troubles me about it.

I feel very confident that I'm getting  the right data, the right analysis on it and the right results.

Dan Wright: [00:09:14] Yeah, that's right. And that's important because , what we've seen is, at an executive level, and , with business analysts, others, they understand that this is a massive sea change that is going on in technology. , there's all sorts of stats out there. There's a recent IDC.

A report that found that spending on AI systems is going to increase by 31% in 2020 from 2019. And PWC forecasts that it could contribute 15 point 7 trillion to the global economy by 2030. So the value is, is very clearly there, but to unlock that value, especially for many of the largest banks, the largest healthcare, institutions and other, , heavily regulated entities, they need to know that what's going on under the hood.

is, , fully auditable, in that they can actually trust their AI. And so we built that in, from the very beginning and I think is one thing that differentiates, differentiates us out there in the

Steve Hamm: [00:10:12] , I think about C-suite people. I mean, I I'm sure they're not your core user, but do do CEO's CEO's are they able to use the tool as well?

Dan Wright: [00:10:23] Yeah, they can absolutely use it. And, we are seeing a lot of demand for that. So I would say because of how strategic this is for companies where they realize that they're sitting on. more data than they've ever had in the past. and that there's tremendous value in that data, but they need some way of automating, , the application of data science to that data in order to unlock the value that there's tremendous demand.

and now what we're able to do is show them, the value that they're getting in a pretty unique way. So we have dashboards that show, what are you actually. getting in terms of ROI, based on your different AI use cases and, that facilitates, , one, I'm gaining a lot of momentum in these companies, because again, once people see the value that they're getting.

It becomes, just very exciting. They want to do more. And then that speeds up adoption. And it also, for us, just allows us to have great conversations with our customers. , just earlier this week, I was talking to, one of the most senior executives at, , one of the largest healthcare companies in the world.

I also talked to another senior executive at one of the largest banks in the world. And we're having strategic conversations with them about, , Hey, they want to become an AI driven enterprise and they don't want to go after a single use case any more. They want to go big and have this really be how they operate their companies going forward.

And the visibility that we provide is a key part of that.

Steve Hamm: [00:11:51]  earlier this year, data robot announced some major new capabilities for the platform. Let's explore a couple of those, which I find particularly intriguing. If you would talk about automated, deep learning.

Dan Wright: [00:12:05] Yeah, this is a really interesting topic. And I think an exciting. A new addition to our platform. So, , first I think when you think about deep learning, it's important to understand what is the difference between deep learning and machine learning. And it really comes down to how the data is presented.

with machine learning, the algorithms pretty much always requires structured data. Well, deep learning networks rely on layers of what we call artificial neural networks. and so it, it just enables,  much more rapid and, and, natural, um, insights. And so  with this new platform, we're really unlocking some pretty exciting new applications of the technology, that, , never previously been possible.

And, we did that by, , as we've kind of taken the approach in the past, leveraging the best of,  open source technology. but also, , applying a kind of enterprise, curation and the security, that we, we built into the platform, to really help customers. And so that's been, really well received by our customers so far.

And we're pretty excited about.  where that's going. And, um,  , we've now got this new suite of deep learning approaches that are a hundred percent ready to, deploy in production. And we built them again from the ground up using, this architecture that improves performance and accuracy.

And we've managed to improve a top model accuracy on 8% of, of datasets, for example, in our repository of internal benchmarks. So, , again, More automation, more accurate models. And, uh,  , that adds up to value for customers.

Steve Hamm: [00:13:47] I think it might be helpful if you differentiate between, machine learning and deep learning, because  there's a significant difference and I think not everybody understands it. So could you do that?

Dan Wright: [00:14:01] I think that the biggest thing to understand with the difference between deep learning and machine learning is, machine learning has been around for a while. So I mentioned that,  even before data robot, pre 2012, there was, uh, machine learning.

It was just done manually. Deep learning is, sort of an emerging, um,  , a field more recently than, than machine learning. And some of the barriers around it have been, the same as machine learning. So having, highly manual process in order to apply deep learning. but then also it's incredibly expensive and time consuming, I would say, even more so.

then, just other, other, applications of machine learning. And now what we've been able to do is we've been able to make, deep learning models, easy and simple. So we significantly boosted our deep learning capabilities. We powered this by a new, Kara space model framework. From which did a robot recently secured a provisional patent.

And, we leveraged, open source projects, like TensorFlow and others and created a new suite of, deep learning approaches that are a hundred percent ready. To deploy in production. So I would say,  the biggest difference with deep learning, is again, it's, it's sort of a newer, emerging area.

And in the past, I would say even more than, machine learning, deep learning has been practiced by a very small subset of the population. And now. We're enabling, , a larger part of the, population to use deep learning by just simplifying and automating the process, which again is core to our

Steve Hamm: [00:15:38] Yeah. Yeah. I think , that's helpful now, visual AI, talk about your new capabilities there.

Dan Wright: [00:15:45] Yeah, I'm really excited about this one and the, the. Headline that I would say for, for this one is, it unlocks so many applications in so many industries for so many different use cases. And I'll give a few examples of it.  first what it is at the, at the top level is,  we can literally take any image and , provide, the application of machine learning to that, that image.

And , how does that actually translate in practice? Well, Retailers can use computer vision to improve customer experience detect when a product is out of stock , on store shelves, and even watch versus suspicious activity to help, with, , theft or, or loss prevention.

Manufacturers can use visual AI to identify product defects in real time. So rather than having to wait and,  , in a, a manufacturer ships, a product, and then,  they found out there's some defect and then you've got to do some massive recall, which can cost millions of dollars. You can be alerted to that.

Before that, release ever happens and avoid just an incredible amount of costs. we do that by, , monitoring all of the parts and components as they come off the production line and then feeding those images into the model to flag potential defects and avoid problems further downstream.

Steve Hamm: [00:17:05] Oh, that's really interesting. .

Dan Wright: [00:17:07] And I can give a couple more examples

Steve Hamm: [00:17:09] I think it would be great if you would. Yeah.

Dan Wright: [00:17:12] one, that might be particularly interesting just given what's going on right now. And, and I found very interesting as with health care. So healthcare providers can use image based neural networks to automate the examination and diagnosis of health issues.

So from MRIs, cat scans, x-rays all of that had to be done by humans in the past, uh, using the human eye. We're now able to apply AI to that analysis, which results in it being much faster and with a much lower error rate. so you can understand, I think at a really human level, the impact and the power of

that technology

Steve Hamm: [00:17:51] I remember not too many years ago when the big new innovation here was to send the MRIs and the other images to India overnight and have, people there examine them and send the results back. But now you can do it with AI probably in seconds rather than, an overnight process.

Dan Wright: [00:18:09] That's right.

Steve Hamm: [00:18:10] that is very, very cool. Now there are a lot of AI companies out there. And it seems like there's a race going on among them to broaden their platforms and become one stop shops for AI based analytics. And you mentioned at the top that you've made several acquisitions there in the past month, how do you see things going?

Is it going to be a lot of best of breed or do you see that you think they're going to be a lot of, kind of a one stop shop companies out there?

Dan Wright: [00:18:39] Yeah.  it's a really big opportunity. When you think about AI, , I mentioned the PWC stat. 15 point 7 trillion is estimated to be contributed to the global economy by 2030. So when you have a market opportunity, that's that big. and it's very clearly here.

It's going to be a noisy market. And of course this one is what we see in the market is that there's a lot of companies that are either saying that they do AI and really what they do is, some very small kind of niche application of machine learning as part of their product, , there's others that.

Are,  AI companies, but they're really focusing on automating one piece of that journey from,  data to value. And so are vision from the very beginning has been to be the system of  record one single platform for enterprise AI, at all steps from the preparation and the tagging of the data all the way to getting models in production.

And then we talked about applications, et cetera, and enabling, these, decisions that enterprises need to make based on their data in an automated way. And I think what we've done now is through both organic development, and I'm always blown away with what our product team continues to release.

, some of the features that we just talked about. And now with, , visual AI and deep learning. Those are great examples of that. but we've been able to push the state of the art and really establish ourselves as the, end to end enterprise AI platform and the leader in the industry.

And I feel really good about where we are in the market. , we were a first mover. We were the first to do that. , I always credit Jeremy's a foresight. , he was one of the top data scientists in the world. He was running. Data science at travelers insurance. We've got a unique vantage point to see this trend coming.

He moved quickly and then the company has been very aggressive.  we've invested hundreds of millions of dollars to creating this technology and bringing it to market. And  now what we think we have is an opportunity now to really accelerate because we have every piece of that end to end platform, and that is really differentiated.

And now we're  in the position where we can, , go into what we call model is AI, which is with these applications and enabling decision-making, for, consumers of AI very rapidly. And there's a very big market for that out there.

Steve Hamm: [00:21:10] it's really sunk into me recently how,  in our economy and the business world, these days, that concentration of power and, , revenue growth and profits really is focusing in on just a few companies.

Like I was looking at, a mutual fund. That was the S and P growth fund. And basically the top 10 investments in that company were all tech companies. I mean in that, in that mutual fund were all tech companies and it was all, I mean, it was like, it was Microsoft, it was Google, it was Facebook, , it was Amazon of course.

And when I think about the world of the future, you could say, , A lot of the profits and a lot of the revenue, a lot of the dynamism in the world is going to come from 15 or 20 companies, not from 500, , it's really, it's a remarkable moment. And, uh, , I also think snowflake has an opportunity to be among the important technology companies.

Dan Wright: [00:22:09] I couldn't agree more, on both of your points. I think we are in a very unique and exciting time. , I think when you look at, the, the fortune 500 and the global 2000. the, the winners for the next generation are being defined, , right. As we speak in technology. And, obviously I, I mentioned that I feel we're well positioned, but I, I totally agree with you on snowflake, just an amazing company. and  I think that the sky is the limit when it comes to snowflake.

Steve Hamm: [00:22:35] Well, I know you're new to the company, but if you could describe kind of the partnership with snowflake, when did it start? What kinds of stuff to the companies do together?

Dan Wright: [00:22:46] So, , I, had been familiar with snowflake obviously before I ever started a data robot. And when I joined. I was very eager to understand where we were in our partnership. Cause I saw a lot of potential there, obviously snowflake, being, the system that more and more customers are using to store and manage their valuable data and data robot, allowing companies to apply data science in an automated way to unlock the value of that data.

You could see the synergy, even as someone who is new and as I've dove in, I think there's just an incredible, amount of potential for the partnership. Snowflake is data robot's number one data platform, partner, and one of our most strategic partners. And, the valuable data that you have, you can now leverage machine learning and give customers a pretty differentiated.

Experience. And so we've really leaned into the partnership. We focused on four different areas in particular. One is on product integration. The second is on customer marketing. The third is on our joint sales go to market, which has been very closely involved in, and then also with the executive team alignment and Jeremy and I have a, , regular contact with, with Frank, but also with, with Chris. and , that partnership is, is really, really good.

Steve Hamm: [00:24:09] So that's Frank Slootman, the CEO of snowflake and Chris Degnan, the chief revenue officer you're referring to.

Dan Wright: [00:24:16] yes. Yeah, exactly.

Steve Hamm: [00:24:18] so get a little granular here. Describe a scenario for a customer using snowflake and data robot together. And, and what kind of value, what kind of value do they get out?

Dan Wright: [00:24:30] Great question. I'll give a, , an example. so one, that I really like is with beacon street services. So beacon street was able to work with data, robot and snowflake to optimize their subscription campaigns. And they're on track now to add 15 million in additional sales.

Directly attributed to their work with snowflake plus data robot. So,  , to me, that's one example, just showing the value and the power of what we can do together, , 15 million in additional sales that will move the needle for almost any company. There's another example that I'll give, which is harmony, harmony, leverages data, robot, and snowflake to maximize the value of its  data science teams.

As a marketplace lending platform, harmony uses both technologies together to gather, understand, and use their critical data in a way that informs their entire customer journey and makes the experience a really delightful for their customers.

Steve Hamm: [00:25:28] So as we speak, we're in the middle of the coven crisis, which is putting incredible stresses on many companies, like never before, how is it affecting data robot and your customers?

Dan Wright: [00:25:43] it's impacting every industry and every business,   , our customers are not immune to that. And we feel that,  we have a unique ability and an obligation really to help them navigate, through their digital transformation, , during this time.

And, , across the board, we're seeing companies that are. Looking for ways to,  empower, there people to be as productive as possible, even as, , the ability to provide additional head count, maybe, more limited  just based on the current environment. And so there's been a massive.

push to automate as many workflows as you possibly can. And also to understand. How should I think about my business decisions now that the world has changed so much. So, whereas before, they already understood that  the world changes rapidly or data changes rapidly. And so you do need to automate the application of AI in order to keep up with those changes.

Now, all of that has been accelerated and  the world is. Changing massively, , from day to day and week to week. And so we're hoping companies not only keep up with that, but actually get in front of that. we're seeing, all of the trends that are happening right now from the changing economy to the shutdown of physical businesses, to the move, to online purchasing, , really accelerate some of our interactions, with our, with our customers.

and , the other thing that we've been trying to do in addition to helping with customers more at a company level is, , we do think it's an opportunity for. AI to, to help at a sort of country and a global level. And so,  we've been able to build a highly accurate model that predicted, 88% of the top 50 counties, most likely for COVID to spread in the U S  , we were, did a blog about that.

We've been able to share with policy makers. , how we see this unfolding and we're trying to communicate very, actively with them so that they can take preemptive measures and, , help citizens take preventative measures as well. We also have launched a dedicated research center to our community, so that they can use data robot.

To perform research that will help with COVID-19 and,  another theme that we've heard from customers is, okay, I've got all my models.  , many companies have hundreds or thousands of, machine learning models in production now. But again, the world has changed.

How do I manage and update my models quickly so that I can continue to get value even in light of that change and make the correct decisions based on that change. Because as we know, if you have a model. And it's not updated. As information changes, you won't get,  accurate predictions. And so we did a webinar called AI in turbulent times where we, talked about best practices and how customers can, , stay up with those changes and make sure that they're getting accurate predictions in the business results that they're looking for.

Steve Hamm: [00:28:44] Now I know you you've offered the platform for use free of charge  for organizations and individuals participating in the Cagle competition, which is sponsored by the white house. This is for COVID related research is that different than the, kind of capabilities you're giving to your regular clients? Or how does that work?

Dan Wright: [00:29:04] so Jeremy and Tom, who co-founded data robot, they were initially very involved with that committee. Unity. And they were actually two of the top ranked data scientists within that community globally. And so it's kind of stayed near and dear to the company. And, , at the same time when we heard that the white house was sponsoring a competition to do COVID related research, And try to help with the current crisis that's going on throughout the country.

We felt like we had to do something. And so   , this was a while back, , right after covert hit, we ended up offering our platform free of charge to anyone who is interested in participating in that competition to help with the response efforts. And that includes both. access to our automated machine learning, but also, access to the  data preparation, uh, solutions that we provide with the idea of being that we could enable all of those, data scientists participating, around the world to get to, , actionable insights much faster than if they had to do it.

Through another means or if they didn't have access to the platform. And so far, the reception has been great. More than 500 people from dozens of countries have signed up for the trial. And these people have spent more than 5,000 hours just on automated machine learning alone.

Steve Hamm: [00:30:21] are you able to see into, , what, what have they accomplished with it, with what the results are?

Or is it too early to tell?

Dan Wright: [00:30:29] we know that they've made a significant amount of, of progress, but I do think that the situation is,  going really rapidly and we're pretty focused right now. That's a big focus for us, not just with this effort, but I'd say as a company, as a whole, doing everything possible.

especially as States look at potentially starting to reopen, to make sure that, , it's done in a way that is I'm going to maximize the safety of everybody involved and also allow,  different counties and States to plan as well as possible when it comes to supplies and, , other things that are going to be important.

And so.  we've made a lot of progress, but I would say the ending to the story is, is still a ways off. And, , we're continuing to do whatever we can every day to

Steve Hamm: [00:31:15] Yeah. . I think that's interesting what you say about counties because clearly. , the crisis doesn't affect every country or every region of every country, the same. And in order to deal with it and, and no proceed sustainably and not have flare ups, it will have to go down to the County to make.

Kind of decisions about what do we do here and that if there's a quick response, negative response, how do we clap back? So a lot of it is going to be this very close management of the real world, and it will require real time data analytics to get it done. I think that's, this may be the most powerful demonstration of, of data analytics and their usefulness.

And AI is usefulness that we've seen.

Dan Wright: [00:32:02] I totally agree. And , it's because it's such a complex problem, right. As you said, it's not at a country level or a state level, it's down to a County level. And so, , on top of that, people move around. Right. And, that's happening even more now. And so the ability to. provide,  automated, processing of the massive amounts of data down to the County level.

And then also we have geospatial and,  as I mentioned, , visually AI built into the platform and that allows you to actually monitor the movement of people and,  create predictions, as that's evolving. So that's why I say, , I do think it's an important moment for AI to, to try to.

Help, during the crisis and we feel an obligation to, , do anything possible to, to help out. So.

Steve Hamm: [00:32:51] Let's look out a few years, five years. What do you think is going to be happening with your company and how do you think data and data analytics will be affecting businesses in society?

Dan Wright: [00:33:03] I mean, there's no doubt that AI is. In my opinion that the most transformative technology of our time, I really believe that the potential for this technology to impact, not only our customers, but the entire world is, really unprecedented. And when I was thinking about,  , what I wanted to do after.

Leaving my last company dynamics and, , where did I want to go next? I was struck by just that massive shift that's happening in the market. , I think it's, it's almost cliche to say right now, but , software ate the world is Mark Andreessen said during the last decade. And I really believe that there's this new trend that's going to happen in 2020 and beyond, which is.

About the adoption of AI in companies, that previously relied on software to digitally transform now taking the next step and racing to adopt AI, to solve their most critical problems and try to gain a, an edge on the competition. and so I like to say AI is, is eating software. In some ways, all those large software companies that came up are now going to raise to adopt AI.

And I think that the potential for that is technology and, or is massive. And, just to give, , some flavor of that, it's very apparent to me now having been with the company sometime that every company is now an AI company. And that includes a, a major retailer like Kroger, which is a large customer of ours too.

, bank like PNC or a healthcare provider, like, , steward health, every single customer across every single industry. Can you get tremendous amounts of value from this technology? and, , once all these models are created and,  they're in production, then you have to have a way of managing them all.

And so I think we're uniquely positioned to not only help with the adoption of AI, but the management of all those models and all of that data, especially when you combine, , data robot with a partner like snowflake. So. I see, I would say a very large market already expanding rapidly over the next five years.

And I do feel like we're in a position to emerge as the leader in that market. And, , that's one reason that I joined and it's, , very excited, exciting for me when I look into

Steve Hamm: [00:35:18] Yeah, it's amazing to think the libraries of the future might be AI models. , I mean, with, with literally millions and millions of models, kind of sitting on the shelf, waiting to be used.

Dan Wright: [00:35:32] Yeah. I mean, that's right. And I think, , the beauty of that is in my mind, it allows for a lot of. freedom for humans to do, the,  sort of highest level work that they can do, and also frees up,  mental space for creativity. Right. I think we've all felt at a personal level. I know I have that.

There's just so much data nowadays. And how do you even manage it all? And  for us, the real opportunity is. even at a personal level, allowing people to automate some of the more manual tasks and just the decision making, that can result in, in a lot of fatigue and inefficiency and enabling people and companies to use those insights, which they generate automatically, to, to make better decisions, which will then free them up to think about, okay, what else is possible

Steve Hamm: [00:36:26] Yeah, really interesting. When you talk about AI eating software. So there's, there might be much less of the, kind of the Drudge work of coding to be done. And that frees up people with, with brains and creativity to be analysts,  and to be inventors. So it really does. I mean, it's, it's a tremendous flipping , from  essentially manual labor into automated labor.

And I think that's really gonna make a huge difference as we go forward.

Dan Wright: [00:36:55] Yeah, exactly. We, we talk a lot about, empowering human and machine intelligence. There's things that, AI, I can do that, frankly. People don't want to do and are best positioned to do. And we can automate a lot of that. And then what it enables and empowers people to do is things that humans are, a better at and B they enjoy doing more.

So I think if you can do that, it really paints a picture of a bright future in front of

Steve Hamm: [00:37:22] Yeah, well, people worry about, , the machines taking over, but I think a different and more positive. And I think actually more realistic model is of collaboration and kind of division of labor in a sense, just like what you're saying.

Dan Wright: [00:37:37] Yeah, I agree. I think that there's a, for whatever reason, maybe it's,  , some of the movies that have come out, but they're, there does seem to be a fear of, of AI and machine learning, , in places and, uh, did that can actually slow people down when it comes to adopting the technology.

But. what we've seen is that there's just tremendous value. If you leverage the best of machine intelligence and combine it with the best of human intelligence, it really pushes the boundary of what's possible. And the applications again are across every industry they're truly global. And we try to reinforce that at every opportunity,  , we even have, our mascot is a very friendly looking.

Cartoon a mascot, which my kids love. They love their data robot t-shirts and, , it's just reinforcing that point that, this is actually something that is empowering to humans. And, we see a future where, humans and, and machines work together in a way that, is, very, empowering to people and enables them to do their highest level work.

Steve Hamm: [00:38:39] So, Dan, thanks so much for your time today, , your stories and insights about what you do with AI and what you do with data, what your customers do, what you enable them to do. It's really been fascinating. And in fact, really exciting for me.

Dan Wright: [00:38:56] Well, thanks so much. I've really enjoyed the conversation and we have, we're very excited to, , continue forward and obviously very excited about our partnership with snowflake. So thank you again for the time